Hi Andreas. In the kindle version of the book, I think you mentioned the images and graphs are online for easier viewing. I can’t find them…can you point me to them? Thanks!
Hi Andreas, the site you referenced for book images have been moved, and I’m not able to find the images in its new website. Can you repost please?
Thanks.
Hi, i will like to subscribe to the equity momentum report. May i know how much is the minimum amount to properly invest in this strategy for a private investor?
It’s possible to implement with quite small amounts. I’ve been running an experiment where I tried it on a tiny account of 10k, and that works fine. Of course, for small accounts you can’t do as exact position sizing and such, but it doesn’t have a significant impact.
Just finished reading the book. Very interesting and kudos for writing it. An eye opener and I seriously doubt I can be same investor again.
I have one question though which was not answered in your book. It appears that if we follow the strategy as detailed in the book, we will most likely end up buying at or near the top range and the stocks may be due for a correction because to reach at the top rankings the stock has to perform well during immediate past and we all know that after a move, generally stocks either correct or take a pause.
Sometimes you end up buying into a strong short term move, sometimes you happen to buy after a short term dip (with long to medium term strength). This strategy does not attempt any sort of short term timing, and it works well anyhow.
Take what you like from the strategy and add your own flavors. If you prefer to use the strategy to find target stocks, and then employ your own short term timing strategy to find the entry points, go for it. Perhaps you can find ways to improve the results.
First of all – big congratulations! This has been (as well as your other book) one of the best and for for sure the most honest trading book on the market.
I have had similar thoughts that Vinay (above) and basically was thinking of limiting the drawdown – I would not be able to bear anything above 20% :-). I have added short term filter, which attempts to buy dips (buy low) and sell high. This means that I would use the ranking as it is described in the book but buy only these stocks that give me the entry signal based on the filter. It seems to work ok and smooth the equity curve for me. Of course as a result I will limit the lag but at the cost of the whipsaw – it is obvious. I understand that for book purposes the strategy described had to be relatively simple, but I was wondering if you have experimented with such additions to the strategy and I wonder if you think it make sense to go this avenue? Thank a lot for your thoughts….
Secondly, you mentioned somewhere in your book that it is not good to keep the account on cash during bear markets and you have mentioned money markets… Given the fact that recent bonds returns are well below inflation rates and CDs are even worse this is becoming tricky. Would you be so kind to shed some light on the preferred alternatives? This would have a lot of value….
All The Best,
Thanks, Tomzo! Would be great if you want to write a review on Amazon.
Drawdowns: If you can’t take a DD of 20%, frankly you shouldn’t be in the equity markets. You probably shouldn’t invest at all then, at least not outside of fixed income and money market space.
Expect realistically that your real life drawdowns will be more than double of your long term annualized returns. If your absolute maximum allowed DD is 20%, you should target a return of 5-6% to be on the safe side. At that rate, you’d have to ask yourself if it’s worse the trouble.
My website places you in Spain. So compare those 5-6% a year with what your government gives you. The 10 year benchmark yield is at the moment 1.8%. Three years ago it was at 7%. It’s now low, since we now know that this yield is pretty much de facto guaranteed by Germany. Seems like a valid alternative.
But of course, I have done work with smoothing strategies. I use some, but of quite simple nature. The problem is that they make you miss a few great positions and that can cost you more than they end up paying for.
Safeguarding cash: This becomes very important if you’re a professional or if you’ve got more cash than your local government depositor guarantee. Cash at a broker or bank can go poof in a second if your custodian goes belly up. Even if it doesn’t give any profits, placing money in treasuries or money market is a way of safeguarding it. You’ll get your instruments back in a crash, but not the cash. This is a defensive strategy only.
Then again, if you’re happy with taking a little risk with the cash to make money, there are alternatives. The most interesting space at the moment is in merchant financing. Very high yields at very low risk. Quite an attractive business, and we’re very active there, expanding fast. As always, I’m very careful not to mix my book website with the actual asset management business, and I don’t want to market any investment products here. Of course, if any qualified investors should contact me directly, that wouldn’t be my fault… 🙂
Great to see your reply – another reason for leaving a reply on Amazon – I will no forget to do so….
I live on Canary Islands and in the UK. Thanks for the explanation on the staying in cash – makes sense. By the way I saw in your bibliography Antonacci’s Dual Momentum Investing book. I was actually thinking of applying his methodology when in bear markets i.e.: investing in aggregate bonds – by the way – does the strategy described in this book makes sense to you? I would really value your opinion…. I am an average trader, but I am a very bad investor – i.e.: I totally ignored this part besides some real estate investments :-).
Regarding the draw downs. I really value what you are saying here and I am trying to understand this. I have been trading markets for nine years and for last four years I have been doing this for a living. I have not been a momentum (neither relative strength) trader though. I have been a swing trader and I am mostly false breakout guy (rather than momentum). I have been trading almost exclusively US equities with average trade time horizon between 3 – 7 days. I do not think I am a good trader at all, but my highest DD on equity curve was 16% and profits though not great were several times higher than 6%. The reason for getting interested in your strategy has been mainly the fact that it is relatively less time consuming and requires attention once a week, whereas my method glues me to the screens for 2 hours every day. It is not much, but I want now to do other stuff in my life, so from that perspective the method shown in your book would be ideal for me.
Getting back to DD – it is of course relative what is high or not ( I am probably very risk averse person), but nevertheless I am wondering why this difference? Is it because my method is shorter time frame and “your” method in more mid or even long-term? My risk on single position, by the way is max 1.5% of ATR (and I don’t trade stocks with ATR higher then 6% of the price to limit the volatility) and the max risk of any trade in my account is 1%. Sorry for a long message, but your info on the DD really has shaken me – thanks in advance for sharing your thoughts, Andreas….
What I mean is that a drawdown of 2-3 times your annualized gain is likely to happen sooner or later. At least, for most trading strategies and for most traders that happens. Not every year, and not even every second year, but it’s something every trader should be prepared for and be able to survive.
Sure, I know people who traded successfully for decades without such big drawdowns, but it’s highly unusual. The ones with the least drawdown are usually the really skilled speed traders, and those are much more rare than pop culture wants you to believe. Most are former investment bank prop traders.
I don’t really think a 25% DD a couple of times per decade is a big deal though. Not if you compound at 10-15% in the end.
Also, as you might have noticed, I try to err on the responsible side. Far too many trading book authors have this attitude that everything is easy, everyone should risk all their money, preferably buying their trading systems and mentoring. Just grab the free money in the markets before anyone else does! The problem is, these guys, with practically no exception have never done this themselves. They just sell get-rich-quick dreams to retail traders, betting on that there will be new suckers after the first ones blow up their accounts.
I’d rather discourage people from taking risks than contribute to people losing their hard earned cash. But if you’ve been doing this so long and with good results, you probably know what you’re doing.
Just keep the old expression about pennies in front of steam rollers in mind. That’s something I see often, even in the business. You find a way of getting small high probability gains. So you keep repeating and repeating, getting tiny gains each time, but it builds up. But what they don’t realize (or don’t care about if it’s OPM) is that while they have a very high probability of a small gain, they have a very low probability of a large loss. Those types of trading strategies tend to go up like Madoff’s return curve for a few years, and then suddenly take a massive hit and lose it all.
If you’re doing something that works, keep doing it. Just be mindful of hidden risks.
I bought your book, red it while I was on holiday and really loved it. I was excited at every page turn. Currently having a decent portfolio of value orientated stock I see in your method a way of diversifying more and exploring new grounds. Willing to give it the necessary effort I started making lists/plans with what I need to investigate and/or consider when I found myself facing an issue. I know that you take into account brokerage fees (20$ per transaction) but what about taxes ? Living in Belgium, from the 1st January of 2016 we will have to pay 33% tax on the plus-values of stocks if you sell within 6 months (without the possibility to deduct min-values) and 27% on dividends (25% before). Although this could possibly be solved by opening an account abroad I would prefer to keep my account in Belgium. My question for you, if you would be so kind to answer, is if you took taxes into consideration and if yes which rate ?
No taxes taken into account, Armin. The tax situation is different for each reader, so there’s no way to make a fair representation. Here in Switzerland there are no capital gains taxes for normal investors. Institutions/funds normally don’t pay such taxes either.
I’m not sure your Belgian tax authority would appreciate the solution of opening an undeclared account abroad, Armin…
If capital gains taxes are a concern, lower the trading frequency. Less frequent rebalancing and higher threshold for position size rebalance. It doesn’t make a big impact on performance.
Also, if you’d like to write an Amazon review, that would be very appreciated, Armin.
Your books are the best trading books I have ever read! Only wished you could have wrote them earlier, so I did not have to go through the massive amounts of nonsense out there before finding this gem. Please continue to write more books. The logical explanations backed by statistical evidence in your approach to writing books clearly stands out from the crowd. Loved your yearly walkthrough!
Do you have the results of running this equity momentum strategy together with the trend following strategy? At the first glance, it seems like both strategies are not highly correlated and combining them would give a better risk adjusted return. More so if the trend following asset universe have lesser equities. How would the results be like if you had no equities in the trend following strategy and used the equity momentum strategy for exposure to equities instead?
It occurs to me that one does not have to use an existing index to follow your strategy. I can take any number of stocks or ETFs to create my own index and then follow your strategy. For example, I can identify any number (say, 30-40) ETFs that reasonably cover all/most market sectors (equities such as large cap-growth, large-cap value, mid-cap growth, mid-cap value, etc. bonds, real estate and commodities) and then add together their end-of-day price quote to create a custom index. With this, I can then develop the mathematical calculations identified in your book to determine the best investment options. I imagine that hedge funds likely do something similar with their favorite stocks.
The theory is straight-forward, although it does require a fair amount of programming or mechanical effort to develop and maintain.
Thank you for writing this book! It was very informative and I plan on implementing it for part of my portfolio. You mentioned in the book that it would be impractical to create the ranking table from Excel for all 500 stocks. I have created one stock recently in Excel just to see how it works. What programs/software/data providers, etc. can you recommend using in order to make the process of creating the ranking table more easily?
Mr. Clenow offers a subscription service which provides the equity ranking table and data described in his book. It is available for $15 per month. Here’s the link:
I am already subscribed to the subscription service. It is very helpful and plan on using it for the S&P 500 index. However, I would like to add in more mid-cap and small-cap options and would like to run similar reports on the S&P 600 & S&P 400.
P.S. How is ATR being calculated on the report? In the book, Andreas mentions using a 20-day ATR. When I look up 20 day ATR using a couple of different charting software packages, I am getting a different number.
Clenow did not provide detailed calculations of the Average True Range (ATR) in his book. He says only, “It’s a common measure of how much an instrument tends to move on an average day, up or down”.
If his on-line calculations are not repeatable, then Clenow owes readers an explanation.
There are a few different variations of ATR, all of them more or less pointless attempts for someone to become famous by making a slight variation of an old concept, which is true for most technical indicators. It really doesn’t matter, as long as you get a close enough value. If you’re reading up about Wilder smoothing versus exponential smoothing versus simple smoothing versus weighted smoothing, then you’re already getting far too bogged down in meaningless details.
The broad concept is what’s important, not the decimals on your calculations.
Also, bear in mind that my numbers are based on total return series, not price series like most charting packages would show. The daily bars I used are constructed from in-house tick level data, and therefore could have tiny differences to official daily OHCL data reported from exchanges. You’ll likely see some rounding errors, but nothing that matters.
Again, it doesn’t really matter. Any trading method that relies on some specific smoothing method and other tiny details just isn’t robust. Trade broad concepts, not over optimized details.
Hi Andreas,
As part of Equity Momentum Trader subscription within the Equity Momentum Performance and Portfolio section you’ve displayed a Current Risk Level gauge 0 – 100, could you please explain what this gauge is based on and its use.
The guage shows the current exposure level. That is, how many percent invested is the portfolio at the moment. The remaining balance being cash of course.
Remember how the model scales exposure up and down depending on the overall market filter? The guage tells you where we are at the moment, how many percent of the total portfolio that is allocated.
I just read Stocks on the Move and absolutely loved it. The risk parity approach at the end was very interesting. I noticed significant under-performance in early 1999 though. Do you know how the approach worked in the 80s and 90s? Equal weighting has significantly out-performed since 2000, but I’ve read that it under performs in the 90s. Do you know if the risk parity approach under-performed as well?
Briefly, last year wasn’t great for momentum stocks and the strategy ended at -0.2% while the S&P Total Return Index closed at +1.3%. It was a tough year with that sudden dip in August, but we ended on equal footings with the index.
Very nice book.
In Chapter 14, you mentioned ” holding a portfolio of 500 stocks may be a little impractical. Simply by using a risk parity weighting, we can …”. So have you tested the test of holding S&P500 (e.g. SPY) and then add one of your rule (Chapter 10) buy SPY if it is over 200 day moving range”.
Just putting a trend filter on the SPY? Sure, that’ll give you much better returns than a long term buy and hold of course. But by picking strong stocks and using a more rational weighting model, you’ll get even better results.
Hi Andreas
As far as I can see there are three main factors to the outperformance of this strategy with respect to the S&P I) the trend filter 2) the high momentum stocks and most interestingly the 3) risk parity approach to position weighting.
With regard to the latter I have 2 questions
i) does this work simply by reducing the contribution of the high market cap stocks – I presume that this is why a random portfolio works better than the S&P. I note that the RSP equal weighted top 500 index has done worse over the last year or so but this could just be cyclical I suppose
ii) is it the rebalancing that makes the difference. Also, another question regarding rebalancing – clearly this implies on a weekly basis buying more of the selected stocks that have relatively underperformed and these might continue to deteriorate – is this an issue or is this actually beneficial. ie why does position rebalancing work – is the selling of the high performing stocks or the buying of the underperforming stocks.
Yes, the cap weighted vs equal weighting can turn out a bit different on a year by year basis, but over time the equal weighting tends to outperform. If you do vola parity weighting, the outperformance will be even larger. Rebase the RSP and SPY to when the former was launched in 2003 and see the substantial difference.
Rebalance: While equal weighting usually means buying the stocks that underperformed, vola parity weighting doesn’t have to. I’m not actually a fan of equal weighting, but it’s still better than market cap weighting.
The reason that rebalancing is necessary is that otherwise you would over time end up with completely random positions. If we assume that you picked your position sizing for a valid reason, you probably want to maintain the same basis. It doesn’t have to be on a weekly basis, but at some interval you need to reset to avoid being all over the map after some time.
1) Historical, Adjusted, good quality price data, including delisted stocks – got that
2) Historical Index Price Data – got that
3) Historical Index Constituents by date – cant find it?
Anyone got any idea how to get the constituents of an index on a given date? Without that data any backtesting would be subject to survivorship bias.
I see new section ‘Experimental Equity Models’. Is this section available to users who subscribe only to equity momentum report and not future intelligence report.
The experimental equity section is open to all subscribers, both equity report and futures report. After some testing, this new technical approach seems to work fine, so I’m about to remove the experimental label and go live with it.
I am having trouble downloading the images from the book. I am getting error messages about the images not being there to download. Do you know if there is a problem with them. I cannot read the excel formulas in the book (kindle version).
Trying to replicate Futures Core strategy a question emerged. In order to handle non-USD futures can you recommend data providers for historical currency spot prices? I’ve looked OANDA’s website and they seem to be quite expensive.
Hello Andreas: In the member section, you show results of momentum and absolute momentum strategies applied to various indexes. Unless I’m misinterpreting the results, it seems like applying strategies to large cap index beat the pants off mid and small cap indexes during test period. Do you find that surprising? Greg
Sorry if I missed this in the reading. When does a stock become eligible again for purchase if there is a 15% gap? I assume the ban isn’t for the life of the company.
For the book, I used a 90 day lookback. Keep in mind that this is just a comfort rule. The sims look just as good without it, but I prefer to have it in to get rid of some scary looking situations.
Your momentum strategy is well-thought out and easy to understand. Thank you.
My only issue is I do not want to trade weekly. Somewhere on this site I read that weekly trading results in 300-400 trades per year. That’s a lot. Even monthly trading is too frequent for me. What can be accomplished trading quarterly?
Great book! It occurred to me that focusing on a single index is risky.
In your book, you use the S&P 500 index and its components to identify momentum stocks in which to invest. But there are several instances where the strategy has you sitting in cash, with no earnings potential. Shouldn’t you then consider another index?
Back in 2008/2009, pretty much every asset class turned to positive correlation, with negative results, except for bonds. Shouldn’t the strategy have then transitioned to a bond ETF and its components?
I am thinking that the starting point of the strategy should be to analyze a broad range of asset classes first, and then identify that class (or top classes) that exhibit the highest positive momentum. These then are the set classes in which to use your strategy.
Even if we have a reoccurrence of 2008, some asset somewhere will have positive momentum, and there is an opportunity to make money.
The issue in equity world is always the beta. You always find yourself a prisoner of the overall market. Correlations are high and while you can get some diversification, you’re still always very dependent on market regime. You could pick other indexes, but the diversification effect is limited. Trading the SPX and the NDX for instance won’t make much of a difference. Trading the SPX and the N225 stocks is of course a different matter.
As for other asset classes, sure. But I prefer to view that as a different strategy. Let me explain that…
My preference is to keep each strategy clean and simple, rather than to make a complex combination. I can still trade many strategies, but keeping them separate makes it easier to evaluate each component. So of course, it does make sense to trade other things in a bear market, but for me that’s a separate strategy.
If you’re looking for cross asset strategies, futures make more sense than ETFs. You’ll get better control, more possibilities and a whole different leverage potential. Bond ETFs for instance are often very low volatility vehicles and you need to put up the cash for it. For futures, you can scale according to risk and let the notional land where it may.
But you’re absolutely right of course. In 2008, the once who cleaned up were the trend following futures. And not by being short equities, but rather on oil, gold, dollar and bonds.
Your book is absolutely the best on the market pertaining to momentum trading. Your thoughts are very clear and concise. I recommended this book many times.
I really would like to see more books that have this caliber and hope you publish more.
After I read your book I had four questions:
1) How did you calculate the return of the strategy.
let r be the return of asset a, s , d ,f on day i.
is it (ai + si + di + fi)/4 and than you cumulate the returns to the end of the investment period?
2) You describe the ATR to calculate positioning size. The formula you describe does not account for the overall size of cash + invested capital, meaning that sometimes the ATR positioning recommends a quantity of stocks that can’t be realised with the current cash. If you are invested in many stocks this can be very frustrating.
How do you solve this problem?
3) Which alternative volatility measures would you recommend ? Is historical daily volatility measured with the standard deviation an alternative?
4) Do you have any experience with filters based on EV or Equity value?
Again don’t misunderstand might questions, I really know how much hard work you put in your work and that it is reallly very clean. I am actually only curios.
Your book really is a eye opener. And – Please write more books 🙂
Yes, that site was a bit of a mess. Believe it or not, but the bloody site was hacked by some freaking ISIS supporter, and one day I found it all black with white Arabic text on it. Apparently that’s a thing that they do. Once it was cleaned up, it took a long time to get get the site removed from various databases of malicious sites. In the end, I decided to just take that site down and move everything here, to a bit more hardened server.
I also see now that I missed replying to your previous questions. I try to reply to comments on the site, but sometimes I miss a few.
1. Calculating the returns: I used RightEdge, a portfolio simulation engine based on CLR.
2. Yes, certainly the position sizing mechanism is simplified. I recently published a more complex method in the premium section, fairly similar to what modern index methodologies do. Read for instance the methodology of the MSCI USA Momentum Index for some ideas.
3. Sure, standard deviation is perfectly fine. I used ATR to simplify somewhat and to use something easy to explain. Then again, I did make use of R2, so I guess I might as well have used StdDev as well. You won’t find much of a practical difference in the end.
4. I’ve model equity curve filters, but never found value in it. Doesn’t mean that there is no value in it, just that I didn’t find it useful myself.
When rebalancing what do you do when you only can rebalance part of the porfolio before you run out of cash?
Ran in to this problem today when doing my monthly rebalance. System told me to add on most positions but I could only do about 70% of them before running out of cash.
I have the exact same problem too. not sure if the way I did my adjustment was correct. Will appreciate if Andreas or others can give me inputs
Ignoring the cash constrain, I work the rebalancing on paper to find out the weightage of each shares (in percentage) within the portfolio. E.g: Stock A – 5%, Stock B 3%, Stock C 8% etc.
When you add all the percentage up, you will realized that you are over 100%. (which I think explains why we ran out of cash.) Let say it sums up to 110%, and we adjust it back to 100% by calculating the adjustment factor (1/1.11 = 0.900).
Take the adjustment factor and apply to each share’s weightage (e.g stock A weightage will become 4.5%), and work backwards to know the quantity of shares to hold within the portfolio.
I do not know why my question that was written this past weekend was deleted but here it is again. I hope that this does not result in 2 similar posts from me. My apologies if that happens.
I like your momentum indicator. It is very similar to one that I was using, but I have a question about yours. It can happen that your momentum indicator can have a negative value but pass all the other criteria for being a candidate for purchase and sorts into the top n of purchase candidates. It also happens that the momentum indicator of a security that is owned can turn negative. Am I correct in assuming that with a negative momentum indicator, the security should be removed from the list of possible purchase candidates and that if held, it should be sold?
Have you looked at Nasdaq 100 momentum systems. I played around in yahoo and just took the 10 best % performers a year end (even the ones that were added late in the year) and then hold for the next 52 weeks. Seemed to work pretty well. Have you coded something similar to this? And do you think it’d work best starting from January or can arbitrary 52 week look backs dates and holds work well? Also I suspect that getting out if the NDX is below the 200 MA makes sense as well.
I have ordered your book “stocks on the move” and waiting delivery period of 4 weeks.
I am really excited to see your article about why trend following doesn’t work in stocks. I agree about the fact that we are just trading beta and all stocks are highly correlated. Also buying few relatively strong stocks is better idea than large diversification. I myself tested only 4 years of data from 100 stocks (of my country) and found extremely positive correlation. It did not make any sense to choose stocks. It was waste of time. I made a rule to use only index as a buy and sell signal. In this test, i found unbelievable results. I was convinced not to use any stop loss also. Only i had to see the liquidity level in stocks for position sizing. Even if i choose number 1 fundamentally or technically strong stocks and market is down, stock price goes down. So i am convinced only to use index as buy and sell signal. Also high beta+high relative strength stocks gives good results in my test.
I am now testing long term trend following in stocks, buy signal is any breakout in upside lets say 20 days high together with index and sell only when the long term momentum goes down and cross below 65 days low. I am getting positive results. Till today i am positive to this system. How is your opinion on long term trend following in stocks?
About the re balancing approach, instead of monthly , is it better idea to buy only when index shows buy and sell when index shows sell?
Thank you so much for such a super book. It has opened a new door for me and exciting. I am now back testing the index and stocks i follow. The vola parity is like a magic for me. Unbelievable results.
Important thing, what happens if we replace minimum variance in your system in weighting of stocks. I mean CLENOW MOMENTUM AND WEIGHTING BY minimum variance. Does it make any difference? Have you experimented by chance? I need your guidance. For me to make back-testing through min variance by excel takes a lot of time. At least i managed to backest 4 years data using your system in excel successfully with promised results, WOW.
Waiting your reply eagerly.
I have finished reading both of your books. I found them to be truly inspirational.
I have been trading Futures portfolio similar to your Future Managed strategy so quite confident that i will be able to test and implement the same.
Regarding your book Stock on the Move, could you please share the brokerage fees / transaction cost considered per share in your return calculations. I might have missed the same while reading.
Thank you again for writing such insightful books. Looking forward to hearing from you.
Hi Andreas
Although a bit late, just read your book, have one question.
In calculating the position sizing the formula you mentioned in the book is Shares = Account Value * Risk Factor / ATR. Here, when you start off, the account value would be equal to the capital you intend to deploy, however, in the subsequent weeks or whenever the next rebalancing what would be the Account Value ? Would it be the latest Portfolio value ?
Yes, latest. The only relevant value is the current mark-to-market valuation. And to be clear, that includes any open, unrealized pnl as well, of course.
Read your books – really great. I have subscribed to the equity momentum report and started trading it on Thursday – was initially surprised that my portfolio was so different from the website, but then noticed that the SPX has only just crept over the 200 ma. Just one question – is the 15% gap purchase rule also a sell signal? I bought a share on Thursday which rose 18% the next day. Should it be sold entirely when I review in a month’s time or just have its position weight adjusted? Many thanks for the great resource.
Hi Andreas,
after reading the book, I’m trying to apply the strategy. If I choose the 0,1% risk (but even higher) and limit the portfolio to the top 20% of the stocks, currently I can invest no more then 60% of the available capital. What does it means? Is it wise to keep frozen the remaining 40% of the capital, or is there a safe way to use all the available capital? Or I’m making something wrong? Thank you!
Hi Andreas, “Stocks on the Move” was an interesting read! I have spent some time replicating the strategy and I am quite close now. I have come across a few issues where I differ and without boring you with the details I would very much appreciate your input on one of them. When an acquisition has been announced but the company to be acquired remains in the index, the volatility (ATR etc) may stay very low for some extended period of time. As a result – and if it still qualifies to remain in the basket (No 15% Gap etc.) – its weight might increase to relatively high figures. It seems to me that you have added a Max weight Cap (or a min ATR floor) to avoid an “artificially” high weighting due to the low ATR. Is that correct? I might have missed it in the book, but did not find any info on such Cap / Floor…
Just as an example (page 224, Initial portfolio 2012): Goodrich Corp (GR). Acquisition by UTX announced in Sep 2011 but Goodrich remains in S&P 500 until July 2012. Your portfolio weight in Jan 2012 was 11.7% which is high but using the actual 20 day ATR at the time, mine comes in significantly higher. The ATR / Vol was extremely low going into 2012. Thanks!
I’ve been reading your book and I’m on the stock ranking chapter. I have been able to find the Adjusted Slope values with the formula you provided. Once we use that, we see Adjusted Slope values for the stocks everyday for (for example) 250 trading days with a 90 day rollover. How do I get a single value to rank the stocks among those 160 or so values? I tried using the last value for Adjusted Slope, but I get values such as 20, 19 etc. as opposed to 257, 240 that you got. Granted, I’m not using your data and much can change to have Adjusted Slope values to go down, but I’m just checking in for my own sanity that I’m not making any mistakes. Am I supposed to use the last Adjusted Slope Value, the average or the sum?
Generally, the last calculated value would be used. Whether you get 20 or 200 depends on the momentum of the stock in question. Either value is quite possible, as would 0 or -200 be as well.
First, thanks for doing the book. Its very refreshing to have some clear presentation in this area (warts and all).
Since we have some degree of correlation across all stock mkts/indexes is there any reason we wouldn’t rank across the entire stock universe rather than a narrow index (perhaps with some mkt cap filter and secondary geographic/sector diversification)? In other words why would we ever choose a narrow versus a much broader pool stocks (eg a deep global index)? If we’re searching for momentum, why artificially limit our search since price is price? Why wouldn’t we include every stock we could possibly buy in our ranking for portfolio inclusion?
Sure, go ahead. The main issue is in how to backtest such an approach. You’ll need to figure out how your total investment universe would have realistically looked ten or twenty years ago. That’s when it gets really convenient to use an index. By using historical index membership, you’ll have a realistic historical universe. If you use a market cap filter or similar, that’s fine, but for backtest you’ll need to find a way to replicate that. To backtest, you’d need to include all delisted tickers, account for mergers and capital events, get hold of historical shares outstanding data to calculate market cap filter etc. It can be done for sure, but it’s not all that easy.
I guess I’m assuming that we’re dealing with common herd behaviour and it’s traits in the induction of trending prices.
If backtesting over the S&P works for a simple momentum strategy then I’m extrapolating that as nominal evidence to all instruments that are traded by humans in markets of sufficient scale. It seems that the only thing we are relying on is that a small but definite proportion of instruments will trend more strongly than others for a sufficient extent to offset the losers. And, that in an upwards trending mkt, there will always be instruments trending more strongly than others.
This seems to just be the definition of a persisting trend within a trending market and that everything else is an attempt at optimization (and that back testing might not be the best guide for that optimization).
My thoughts would not be to necessarily rank and select stocks directly from some global index but to rank all available indexes/mkts for momentum and then rank instruments within those indexes and take positions accordingly (whilst recognizing the additional benefits of reduced correlation that different indexes might offer).
Of course this is just my easy speculation (ie there may well be specific behavioural aspects to market making herds in different markets that backtesting might unearth).
Very interesting book. I may have missed this in your book but do all your simulation opens and closes occur at end of day close prices on the day signals are triggered?
Following the Trend
Hi Andreas. In the kindle version of the book, I think you mentioned the images and graphs are online for easier viewing. I can’t find them…can you point me to them? Thanks!
Hi Matt,
All the images from the book are here: http://www.followingthetrend.com/stocks-on-the-move-figures-and-charts/
[edited to update link]
Would it also be possible to get the Trade Log that was created by the simulation in the book?
Hi Roger,
I did those backtests, and wrote that book, some 6-7 years ago and I don’t have either the data or the same software anymore.
Hi Andreas, the site you referenced for book images have been moved, and I’m not able to find the images in its new website. Can you repost please?
Thanks.
http://www.followingthetrend.com/stocks-on-the-move-figures-and-charts/
Sorry James. Seems like the redirect is down. I’ll fix the link…
Hi, i will like to subscribe to the equity momentum report. May i know how much is the minimum amount to properly invest in this strategy for a private investor?
It’s possible to implement with quite small amounts. I’ve been running an experiment where I tried it on a tiny account of 10k, and that works fine. Of course, for small accounts you can’t do as exact position sizing and such, but it doesn’t have a significant impact.
Mr. Clenow,
I just finished reading your book, “Stocks on the Move”. Your S&P 500 strategy is very interesting. I have two questions:
1. Do you use limit or market orders when you trade?
2. Can you provide the number of trades during the simulation period 1999 to 2014?
Thank you.
Ed
Just finished reading the book. Very interesting and kudos for writing it. An eye opener and I seriously doubt I can be same investor again.
I have one question though which was not answered in your book. It appears that if we follow the strategy as detailed in the book, we will most likely end up buying at or near the top range and the stocks may be due for a correction because to reach at the top rankings the stock has to perform well during immediate past and we all know that after a move, generally stocks either correct or take a pause.
What is your view on this. Thanks.
Sometimes you end up buying into a strong short term move, sometimes you happen to buy after a short term dip (with long to medium term strength). This strategy does not attempt any sort of short term timing, and it works well anyhow.
Take what you like from the strategy and add your own flavors. If you prefer to use the strategy to find target stocks, and then employ your own short term timing strategy to find the entry points, go for it. Perhaps you can find ways to improve the results.
That makes sense. Thanks.
Hi Andreas,
First of all – big congratulations! This has been (as well as your other book) one of the best and for for sure the most honest trading book on the market.
I have had similar thoughts that Vinay (above) and basically was thinking of limiting the drawdown – I would not be able to bear anything above 20% :-). I have added short term filter, which attempts to buy dips (buy low) and sell high. This means that I would use the ranking as it is described in the book but buy only these stocks that give me the entry signal based on the filter. It seems to work ok and smooth the equity curve for me. Of course as a result I will limit the lag but at the cost of the whipsaw – it is obvious. I understand that for book purposes the strategy described had to be relatively simple, but I was wondering if you have experimented with such additions to the strategy and I wonder if you think it make sense to go this avenue? Thank a lot for your thoughts….
Secondly, you mentioned somewhere in your book that it is not good to keep the account on cash during bear markets and you have mentioned money markets… Given the fact that recent bonds returns are well below inflation rates and CDs are even worse this is becoming tricky. Would you be so kind to shed some light on the preferred alternatives? This would have a lot of value….
All The Best,
Tom
Thanks, Tomzo! Would be great if you want to write a review on Amazon.
Drawdowns: If you can’t take a DD of 20%, frankly you shouldn’t be in the equity markets. You probably shouldn’t invest at all then, at least not outside of fixed income and money market space.
Expect realistically that your real life drawdowns will be more than double of your long term annualized returns. If your absolute maximum allowed DD is 20%, you should target a return of 5-6% to be on the safe side. At that rate, you’d have to ask yourself if it’s worse the trouble.
My website places you in Spain. So compare those 5-6% a year with what your government gives you. The 10 year benchmark yield is at the moment 1.8%. Three years ago it was at 7%. It’s now low, since we now know that this yield is pretty much de facto guaranteed by Germany. Seems like a valid alternative.
But of course, I have done work with smoothing strategies. I use some, but of quite simple nature. The problem is that they make you miss a few great positions and that can cost you more than they end up paying for.
Safeguarding cash: This becomes very important if you’re a professional or if you’ve got more cash than your local government depositor guarantee. Cash at a broker or bank can go poof in a second if your custodian goes belly up. Even if it doesn’t give any profits, placing money in treasuries or money market is a way of safeguarding it. You’ll get your instruments back in a crash, but not the cash. This is a defensive strategy only.
Then again, if you’re happy with taking a little risk with the cash to make money, there are alternatives. The most interesting space at the moment is in merchant financing. Very high yields at very low risk. Quite an attractive business, and we’re very active there, expanding fast. As always, I’m very careful not to mix my book website with the actual asset management business, and I don’t want to market any investment products here. Of course, if any qualified investors should contact me directly, that wouldn’t be my fault… 🙂
Hi Andreas,
Great to see your reply – another reason for leaving a reply on Amazon – I will no forget to do so….
I live on Canary Islands and in the UK. Thanks for the explanation on the staying in cash – makes sense. By the way I saw in your bibliography Antonacci’s Dual Momentum Investing book. I was actually thinking of applying his methodology when in bear markets i.e.: investing in aggregate bonds – by the way – does the strategy described in this book makes sense to you? I would really value your opinion…. I am an average trader, but I am a very bad investor – i.e.: I totally ignored this part besides some real estate investments :-).
Regarding the draw downs. I really value what you are saying here and I am trying to understand this. I have been trading markets for nine years and for last four years I have been doing this for a living. I have not been a momentum (neither relative strength) trader though. I have been a swing trader and I am mostly false breakout guy (rather than momentum). I have been trading almost exclusively US equities with average trade time horizon between 3 – 7 days. I do not think I am a good trader at all, but my highest DD on equity curve was 16% and profits though not great were several times higher than 6%. The reason for getting interested in your strategy has been mainly the fact that it is relatively less time consuming and requires attention once a week, whereas my method glues me to the screens for 2 hours every day. It is not much, but I want now to do other stuff in my life, so from that perspective the method shown in your book would be ideal for me.
Getting back to DD – it is of course relative what is high or not ( I am probably very risk averse person), but nevertheless I am wondering why this difference? Is it because my method is shorter time frame and “your” method in more mid or even long-term? My risk on single position, by the way is max 1.5% of ATR (and I don’t trade stocks with ATR higher then 6% of the price to limit the volatility) and the max risk of any trade in my account is 1%. Sorry for a long message, but your info on the DD really has shaken me – thanks in advance for sharing your thoughts, Andreas….
Best Luck,
Tom
What I mean is that a drawdown of 2-3 times your annualized gain is likely to happen sooner or later. At least, for most trading strategies and for most traders that happens. Not every year, and not even every second year, but it’s something every trader should be prepared for and be able to survive.
Sure, I know people who traded successfully for decades without such big drawdowns, but it’s highly unusual. The ones with the least drawdown are usually the really skilled speed traders, and those are much more rare than pop culture wants you to believe. Most are former investment bank prop traders.
I don’t really think a 25% DD a couple of times per decade is a big deal though. Not if you compound at 10-15% in the end.
Also, as you might have noticed, I try to err on the responsible side. Far too many trading book authors have this attitude that everything is easy, everyone should risk all their money, preferably buying their trading systems and mentoring. Just grab the free money in the markets before anyone else does! The problem is, these guys, with practically no exception have never done this themselves. They just sell get-rich-quick dreams to retail traders, betting on that there will be new suckers after the first ones blow up their accounts.
I’d rather discourage people from taking risks than contribute to people losing their hard earned cash. But if you’ve been doing this so long and with good results, you probably know what you’re doing.
Just keep the old expression about pennies in front of steam rollers in mind. That’s something I see often, even in the business. You find a way of getting small high probability gains. So you keep repeating and repeating, getting tiny gains each time, but it builds up. But what they don’t realize (or don’t care about if it’s OPM) is that while they have a very high probability of a small gain, they have a very low probability of a large loss. Those types of trading strategies tend to go up like Madoff’s return curve for a few years, and then suddenly take a massive hit and lose it all.
If you’re doing something that works, keep doing it. Just be mindful of hidden risks.
Hi Andreas,
I have finished your book. I enjoyed it. I have a few short questions:
1. The 15 % gaps are from yesterday’s close to today’s open or today’s close ?
2. How do you adjust in real time use the back data for dividends so it does not affect past reality taking into account you use log scale ?
Thanks in advance,
George
Hi Andreas,
I bought your book, red it while I was on holiday and really loved it. I was excited at every page turn. Currently having a decent portfolio of value orientated stock I see in your method a way of diversifying more and exploring new grounds. Willing to give it the necessary effort I started making lists/plans with what I need to investigate and/or consider when I found myself facing an issue. I know that you take into account brokerage fees (20$ per transaction) but what about taxes ? Living in Belgium, from the 1st January of 2016 we will have to pay 33% tax on the plus-values of stocks if you sell within 6 months (without the possibility to deduct min-values) and 27% on dividends (25% before). Although this could possibly be solved by opening an account abroad I would prefer to keep my account in Belgium. My question for you, if you would be so kind to answer, is if you took taxes into consideration and if yes which rate ?
Thank you in advance,
Wish you all the best,
Armin
No taxes taken into account, Armin. The tax situation is different for each reader, so there’s no way to make a fair representation. Here in Switzerland there are no capital gains taxes for normal investors. Institutions/funds normally don’t pay such taxes either.
I’m not sure your Belgian tax authority would appreciate the solution of opening an undeclared account abroad, Armin…
If capital gains taxes are a concern, lower the trading frequency. Less frequent rebalancing and higher threshold for position size rebalance. It doesn’t make a big impact on performance.
Also, if you’d like to write an Amazon review, that would be very appreciated, Armin.
Hello Andreas,
Your books are the best trading books I have ever read! Only wished you could have wrote them earlier, so I did not have to go through the massive amounts of nonsense out there before finding this gem. Please continue to write more books. The logical explanations backed by statistical evidence in your approach to writing books clearly stands out from the crowd. Loved your yearly walkthrough!
Do you have the results of running this equity momentum strategy together with the trend following strategy? At the first glance, it seems like both strategies are not highly correlated and combining them would give a better risk adjusted return. More so if the trend following asset universe have lesser equities. How would the results be like if you had no equities in the trend following strategy and used the equity momentum strategy for exposure to equities instead?
Boon
Mr. Clenow,
It occurs to me that one does not have to use an existing index to follow your strategy. I can take any number of stocks or ETFs to create my own index and then follow your strategy. For example, I can identify any number (say, 30-40) ETFs that reasonably cover all/most market sectors (equities such as large cap-growth, large-cap value, mid-cap growth, mid-cap value, etc. bonds, real estate and commodities) and then add together their end-of-day price quote to create a custom index. With this, I can then develop the mathematical calculations identified in your book to determine the best investment options. I imagine that hedge funds likely do something similar with their favorite stocks.
The theory is straight-forward, although it does require a fair amount of programming or mechanical effort to develop and maintain.
Do you agree?
Hello Andreas,
Thank you for writing this book! It was very informative and I plan on implementing it for part of my portfolio. You mentioned in the book that it would be impractical to create the ranking table from Excel for all 500 stocks. I have created one stock recently in Excel just to see how it works. What programs/software/data providers, etc. can you recommend using in order to make the process of creating the ranking table more easily?
Thank you for your time and help.
Best,
Zack
Zack,
Mr. Clenow offers a subscription service which provides the equity ranking table and data described in his book. It is available for $15 per month. Here’s the link:
http://www.followingthetrend.com/equity-momentum-report/
Ed
Hi Ed,
I am already subscribed to the subscription service. It is very helpful and plan on using it for the S&P 500 index. However, I would like to add in more mid-cap and small-cap options and would like to run similar reports on the S&P 600 & S&P 400.
P.S. How is ATR being calculated on the report? In the book, Andreas mentions using a 20-day ATR. When I look up 20 day ATR using a couple of different charting software packages, I am getting a different number.
Thank you for your help!
Best,
Zack
Zack,
Clenow did not provide detailed calculations of the Average True Range (ATR) in his book. He says only, “It’s a common measure of how much an instrument tends to move on an average day, up or down”.
If his on-line calculations are not repeatable, then Clenow owes readers an explanation.
There are a few different variations of ATR, all of them more or less pointless attempts for someone to become famous by making a slight variation of an old concept, which is true for most technical indicators. It really doesn’t matter, as long as you get a close enough value. If you’re reading up about Wilder smoothing versus exponential smoothing versus simple smoothing versus weighted smoothing, then you’re already getting far too bogged down in meaningless details.
The broad concept is what’s important, not the decimals on your calculations.
Also, bear in mind that my numbers are based on total return series, not price series like most charting packages would show. The daily bars I used are constructed from in-house tick level data, and therefore could have tiny differences to official daily OHCL data reported from exchanges. You’ll likely see some rounding errors, but nothing that matters.
Again, it doesn’t really matter. Any trading method that relies on some specific smoothing method and other tiny details just isn’t robust. Trade broad concepts, not over optimized details.
Hi Andreas,
As part of Equity Momentum Trader subscription within the Equity Momentum Performance and Portfolio section you’ve displayed a Current Risk Level gauge 0 – 100, could you please explain what this gauge is based on and its use.
Regards
Glenn
Hi Glenn,
The guage shows the current exposure level. That is, how many percent invested is the portfolio at the moment. The remaining balance being cash of course.
Remember how the model scales exposure up and down depending on the overall market filter? The guage tells you where we are at the moment, how many percent of the total portfolio that is allocated.
Hi Andreas,
I just read Stocks on the Move and absolutely loved it. The risk parity approach at the end was very interesting. I noticed significant under-performance in early 1999 though. Do you know how the approach worked in the 80s and 90s? Equal weighting has significantly out-performed since 2000, but I’ve read that it under performs in the 90s. Do you know if the risk parity approach under-performed as well?
Mr. Clenow,
Will you be publishing the 2015 performance results of the equity momentum strategy?
Thanks,
Ed
The results, portfolio holdings and ranking tables are updated daily in the equity report: http://www.followingthetrend.com/equity-momentum-report/
Briefly, last year wasn’t great for momentum stocks and the strategy ended at -0.2% while the S&P Total Return Index closed at +1.3%. It was a tough year with that sudden dip in August, but we ended on equal footings with the index.
Andreas,
Very nice book.
In Chapter 14, you mentioned ” holding a portfolio of 500 stocks may be a little impractical. Simply by using a risk parity weighting, we can …”. So have you tested the test of holding S&P500 (e.g. SPY) and then add one of your rule (Chapter 10) buy SPY if it is over 200 day moving range”.
Thanks
Just putting a trend filter on the SPY? Sure, that’ll give you much better returns than a long term buy and hold of course. But by picking strong stocks and using a more rational weighting model, you’ll get even better results.
Hi Andreas
As far as I can see there are three main factors to the outperformance of this strategy with respect to the S&P I) the trend filter 2) the high momentum stocks and most interestingly the 3) risk parity approach to position weighting.
With regard to the latter I have 2 questions
i) does this work simply by reducing the contribution of the high market cap stocks – I presume that this is why a random portfolio works better than the S&P. I note that the RSP equal weighted top 500 index has done worse over the last year or so but this could just be cyclical I suppose
ii) is it the rebalancing that makes the difference. Also, another question regarding rebalancing – clearly this implies on a weekly basis buying more of the selected stocks that have relatively underperformed and these might continue to deteriorate – is this an issue or is this actually beneficial. ie why does position rebalancing work – is the selling of the high performing stocks or the buying of the underperforming stocks.
Hi Chris,
Yes, the cap weighted vs equal weighting can turn out a bit different on a year by year basis, but over time the equal weighting tends to outperform. If you do vola parity weighting, the outperformance will be even larger. Rebase the RSP and SPY to when the former was launched in 2003 and see the substantial difference.
Rebalance: While equal weighting usually means buying the stocks that underperformed, vola parity weighting doesn’t have to. I’m not actually a fan of equal weighting, but it’s still better than market cap weighting.
The reason that rebalancing is necessary is that otherwise you would over time end up with completely random positions. If we assume that you picked your position sizing for a valid reason, you probably want to maintain the same basis. It doesn’t have to be on a weekly basis, but at some interval you need to reset to avoid being all over the map after some time.
To be able to back test this strategy you need:
1) Historical, Adjusted, good quality price data, including delisted stocks – got that
2) Historical Index Price Data – got that
3) Historical Index Constituents by date – cant find it?
Anyone got any idea how to get the constituents of an index on a given date? Without that data any backtesting would be subject to survivorship bias.
Thanks!
Hi Sam, if what you need is not too complex I may be able to help. could you drop me an email on chingang.chew@gmail.com on what you need.
Hey Andrew, I’m working on a review of your book. Great stuff. Thanks for writing it. I’ve already added it on our list of “must reads” for momentum investing here: http://blog.alphaarchitect.com/2015/01/03/an-introduction-to-investing-and-how-to-use-our-site/ and we’ll release a full book review soon.
Wes
Hi Wes,
Thank you for that! Much appreciated. I saw you on the list of speakers for QuantCon in April, so I’m looking forward to meeting you there.
ac
Hi Andreas,
I see new section ‘Experimental Equity Models’. Is this section available to users who subscribe only to equity momentum report and not future intelligence report.
Govind
Hi Govind,
The experimental equity section is open to all subscribers, both equity report and futures report. After some testing, this new technical approach seems to work fine, so I’m about to remove the experimental label and go live with it.
Hi Andreas,
I am having trouble downloading the images from the book. I am getting error messages about the images not being there to download. Do you know if there is a problem with them. I cannot read the excel formulas in the book (kindle version).
Thanks, Chris
Hi Chris,
Kindle is a truly horrible format for images. I’ve worked with their support to get the best images quality possible, and it still looks pretty bad.
All the images are available in high resolution here: http://www.followingthetrend.com/stocks-on-the-move-figures-and-charts/
Hi Andreas,
Trying to replicate Futures Core strategy a question emerged. In order to handle non-USD futures can you recommend data providers for historical currency spot prices? I’ve looked OANDA’s website and they seem to be quite expensive.
Thanks and best regards,
Vitaly
Spot currency prices are easy. They’re not subject to any exchange fees.
Get them for free from Yahoo Finance, Google Finance or Quandl or similar.
Andreas
Hello Andreas: In the member section, you show results of momentum and absolute momentum strategies applied to various indexes. Unless I’m misinterpreting the results, it seems like applying strategies to large cap index beat the pants off mid and small cap indexes during test period. Do you find that surprising? Greg
Not too surprising. Momentum investing makes much more sense than traditional index methodology.
Mr. Clenow,
What do you believe is the minimum number of individual equities that should be in a customized index in order to follow your momentum model?
Thank you.
I didn’t model for a minimum portfolio yet. Guessing, I’d say that less than 15 stocks would be an issue.
Hi,
Sorry if I missed this in the reading. When does a stock become eligible again for purchase if there is a 15% gap? I assume the ban isn’t for the life of the company.
Tks!
For the book, I used a 90 day lookback. Keep in mind that this is just a comfort rule. The sims look just as good without it, but I prefer to have it in to get rid of some scary looking situations.
Just to clarify, do you look back 90 trading days or 90 calendar days?
Trading days.
One more thing, is a stock disqualified if it opens at 100 and closes and 116 or is it only if it opens >15% over the previous days close, or both?
Mr. Clenow,
Your momentum strategy is well-thought out and easy to understand. Thank you.
My only issue is I do not want to trade weekly. Somewhere on this site I read that weekly trading results in 300-400 trades per year. That’s a lot. Even monthly trading is too frequent for me. What can be accomplished trading quarterly?
Thanks.
Andreas,
Great book! It occurred to me that focusing on a single index is risky.
In your book, you use the S&P 500 index and its components to identify momentum stocks in which to invest. But there are several instances where the strategy has you sitting in cash, with no earnings potential. Shouldn’t you then consider another index?
Back in 2008/2009, pretty much every asset class turned to positive correlation, with negative results, except for bonds. Shouldn’t the strategy have then transitioned to a bond ETF and its components?
I am thinking that the starting point of the strategy should be to analyze a broad range of asset classes first, and then identify that class (or top classes) that exhibit the highest positive momentum. These then are the set classes in which to use your strategy.
Even if we have a reoccurrence of 2008, some asset somewhere will have positive momentum, and there is an opportunity to make money.
What do you think?
Thanks, Edward!
The issue in equity world is always the beta. You always find yourself a prisoner of the overall market. Correlations are high and while you can get some diversification, you’re still always very dependent on market regime. You could pick other indexes, but the diversification effect is limited. Trading the SPX and the NDX for instance won’t make much of a difference. Trading the SPX and the N225 stocks is of course a different matter.
As for other asset classes, sure. But I prefer to view that as a different strategy. Let me explain that…
My preference is to keep each strategy clean and simple, rather than to make a complex combination. I can still trade many strategies, but keeping them separate makes it easier to evaluate each component. So of course, it does make sense to trade other things in a bear market, but for me that’s a separate strategy.
If you’re looking for cross asset strategies, futures make more sense than ETFs. You’ll get better control, more possibilities and a whole different leverage potential. Bond ETFs for instance are often very low volatility vehicles and you need to put up the cash for it. For futures, you can scale according to risk and let the notional land where it may.
But you’re absolutely right of course. In 2008, the once who cleaned up were the trend following futures. And not by being short equities, but rather on oil, gold, dollar and bonds.
Can someone please post 2016 monthly results for the Equity Momentum model?
Thank you.
Hello Andreas,
Your book is absolutely the best on the market pertaining to momentum trading. Your thoughts are very clear and concise. I recommended this book many times.
I really would like to see more books that have this caliber and hope you publish more.
After I read your book I had four questions:
1) How did you calculate the return of the strategy.
let r be the return of asset a, s , d ,f on day i.
is it (ai + si + di + fi)/4 and than you cumulate the returns to the end of the investment period?
2) You describe the ATR to calculate positioning size. The formula you describe does not account for the overall size of cash + invested capital, meaning that sometimes the ATR positioning recommends a quantity of stocks that can’t be realised with the current cash. If you are invested in many stocks this can be very frustrating.
How do you solve this problem?
3) Which alternative volatility measures would you recommend ? Is historical daily volatility measured with the standard deviation an alternative?
4) Do you have any experience with filters based on EV or Equity value?
Again don’t misunderstand might questions, I really know how much hard work you put in your work and that it is reallly very clean. I am actually only curios.
Your book really is a eye opener. And – Please write more books 🙂
Hello Andreas,
the pictures of your book are not available. The link is down : http://www.stocksonthemove.net/book-images/
Could you please upload them again?
Thanks
Hi Vaasa,
Yes, that site was a bit of a mess. Believe it or not, but the bloody site was hacked by some freaking ISIS supporter, and one day I found it all black with white Arabic text on it. Apparently that’s a thing that they do. Once it was cleaned up, it took a long time to get get the site removed from various databases of malicious sites. In the end, I decided to just take that site down and move everything here, to a bit more hardened server.
The chart images are all here: http://www.followingthetrend.com/stocks-on-the-move-figures-and-charts/
I also see now that I missed replying to your previous questions. I try to reply to comments on the site, but sometimes I miss a few.
1. Calculating the returns: I used RightEdge, a portfolio simulation engine based on CLR.
2. Yes, certainly the position sizing mechanism is simplified. I recently published a more complex method in the premium section, fairly similar to what modern index methodologies do. Read for instance the methodology of the MSCI USA Momentum Index for some ideas.
3. Sure, standard deviation is perfectly fine. I used ATR to simplify somewhat and to use something easy to explain. Then again, I did make use of R2, so I guess I might as well have used StdDev as well. You won’t find much of a practical difference in the end.
4. I’ve model equity curve filters, but never found value in it. Doesn’t mean that there is no value in it, just that I didn’t find it useful myself.
When rebalancing what do you do when you only can rebalance part of the porfolio before you run out of cash?
Ran in to this problem today when doing my monthly rebalance. System told me to add on most positions but I could only do about 70% of them before running out of cash.
I have the exact same problem too. not sure if the way I did my adjustment was correct. Will appreciate if Andreas or others can give me inputs
Ignoring the cash constrain, I work the rebalancing on paper to find out the weightage of each shares (in percentage) within the portfolio. E.g: Stock A – 5%, Stock B 3%, Stock C 8% etc.
When you add all the percentage up, you will realized that you are over 100%. (which I think explains why we ran out of cash.) Let say it sums up to 110%, and we adjust it back to 100% by calculating the adjustment factor (1/1.11 = 0.900).
Take the adjustment factor and apply to each share’s weightage (e.g stock A weightage will become 4.5%), and work backwards to know the quantity of shares to hold within the portfolio.
Mr. Clenow,
I just finished reading your book, “Stocks on the Move”. Your S&P 500 strategy is very interesting. I have two questions:
1. Do you use limit or market orders when you trade?
2. Can you provide the number of trades during the simulation period 1999 to 2014?
Thank you.
Ed
Andreas
I do not know why my question that was written this past weekend was deleted but here it is again. I hope that this does not result in 2 similar posts from me. My apologies if that happens.
I like your momentum indicator. It is very similar to one that I was using, but I have a question about yours. It can happen that your momentum indicator can have a negative value but pass all the other criteria for being a candidate for purchase and sorts into the top n of purchase candidates. It also happens that the momentum indicator of a security that is owned can turn negative. Am I correct in assuming that with a negative momentum indicator, the security should be removed from the list of possible purchase candidates and that if held, it should be sold?
Thank you
Rick
You could set a minimum required slope in the rebalancing process, Rick. That usually has a positive effect on the long term results.
Andreas,
Have you looked at Nasdaq 100 momentum systems. I played around in yahoo and just took the 10 best % performers a year end (even the ones that were added late in the year) and then hold for the next 52 weeks. Seemed to work pretty well. Have you coded something similar to this? And do you think it’d work best starting from January or can arbitrary 52 week look backs dates and holds work well? Also I suspect that getting out if the NDX is below the 200 MA makes sense as well.
Let me know your thoughts. Thank you
HI Andreas,
I have ordered your book “stocks on the move” and waiting delivery period of 4 weeks.
I am really excited to see your article about why trend following doesn’t work in stocks. I agree about the fact that we are just trading beta and all stocks are highly correlated. Also buying few relatively strong stocks is better idea than large diversification. I myself tested only 4 years of data from 100 stocks (of my country) and found extremely positive correlation. It did not make any sense to choose stocks. It was waste of time. I made a rule to use only index as a buy and sell signal. In this test, i found unbelievable results. I was convinced not to use any stop loss also. Only i had to see the liquidity level in stocks for position sizing. Even if i choose number 1 fundamentally or technically strong stocks and market is down, stock price goes down. So i am convinced only to use index as buy and sell signal. Also high beta+high relative strength stocks gives good results in my test.
I am now testing long term trend following in stocks, buy signal is any breakout in upside lets say 20 days high together with index and sell only when the long term momentum goes down and cross below 65 days low. I am getting positive results. Till today i am positive to this system. How is your opinion on long term trend following in stocks?
About the re balancing approach, instead of monthly , is it better idea to buy only when index shows buy and sell when index shows sell?
Dear Andreas,
Thank you so much for such a super book. It has opened a new door for me and exciting. I am now back testing the index and stocks i follow. The vola parity is like a magic for me. Unbelievable results.
Important thing, what happens if we replace minimum variance in your system in weighting of stocks. I mean CLENOW MOMENTUM AND WEIGHTING BY minimum variance. Does it make any difference? Have you experimented by chance? I need your guidance. For me to make back-testing through min variance by excel takes a lot of time. At least i managed to backest 4 years data using your system in excel successfully with promised results, WOW.
Waiting your reply eagerly.
Thank you
Hi Andreas,
I have finished reading both of your books. I found them to be truly inspirational.
I have been trading Futures portfolio similar to your Future Managed strategy so quite confident that i will be able to test and implement the same.
Regarding your book Stock on the Move, could you please share the brokerage fees / transaction cost considered per share in your return calculations. I might have missed the same while reading.
Thank you again for writing such insightful books. Looking forward to hearing from you.
Hi Andreas
Although a bit late, just read your book, have one question.
In calculating the position sizing the formula you mentioned in the book is Shares = Account Value * Risk Factor / ATR. Here, when you start off, the account value would be equal to the capital you intend to deploy, however, in the subsequent weeks or whenever the next rebalancing what would be the Account Value ? Would it be the latest Portfolio value ?
Yes, latest. The only relevant value is the current mark-to-market valuation. And to be clear, that includes any open, unrealized pnl as well, of course.
Read your books – really great. I have subscribed to the equity momentum report and started trading it on Thursday – was initially surprised that my portfolio was so different from the website, but then noticed that the SPX has only just crept over the 200 ma. Just one question – is the 15% gap purchase rule also a sell signal? I bought a share on Thursday which rose 18% the next day. Should it be sold entirely when I review in a month’s time or just have its position weight adjusted? Many thanks for the great resource.
Hi Andreas,
after reading the book, I’m trying to apply the strategy. If I choose the 0,1% risk (but even higher) and limit the portfolio to the top 20% of the stocks, currently I can invest no more then 60% of the available capital. What does it means? Is it wise to keep frozen the remaining 40% of the capital, or is there a safe way to use all the available capital? Or I’m making something wrong? Thank you!
Hi Andreas, “Stocks on the Move” was an interesting read! I have spent some time replicating the strategy and I am quite close now. I have come across a few issues where I differ and without boring you with the details I would very much appreciate your input on one of them. When an acquisition has been announced but the company to be acquired remains in the index, the volatility (ATR etc) may stay very low for some extended period of time. As a result – and if it still qualifies to remain in the basket (No 15% Gap etc.) – its weight might increase to relatively high figures. It seems to me that you have added a Max weight Cap (or a min ATR floor) to avoid an “artificially” high weighting due to the low ATR. Is that correct? I might have missed it in the book, but did not find any info on such Cap / Floor…
Just as an example (page 224, Initial portfolio 2012): Goodrich Corp (GR). Acquisition by UTX announced in Sep 2011 but Goodrich remains in S&P 500 until July 2012. Your portfolio weight in Jan 2012 was 11.7% which is high but using the actual 20 day ATR at the time, mine comes in significantly higher. The ATR / Vol was extremely low going into 2012. Thanks!
Hi Andres,
I’ve been reading your book and I’m on the stock ranking chapter. I have been able to find the Adjusted Slope values with the formula you provided. Once we use that, we see Adjusted Slope values for the stocks everyday for (for example) 250 trading days with a 90 day rollover. How do I get a single value to rank the stocks among those 160 or so values? I tried using the last value for Adjusted Slope, but I get values such as 20, 19 etc. as opposed to 257, 240 that you got. Granted, I’m not using your data and much can change to have Adjusted Slope values to go down, but I’m just checking in for my own sanity that I’m not making any mistakes. Am I supposed to use the last Adjusted Slope Value, the average or the sum?
Thanks,
Imtiaz
Generally, the last calculated value would be used. Whether you get 20 or 200 depends on the momentum of the stock in question. Either value is quite possible, as would 0 or -200 be as well.
Hi Andrea, “Disqualify any stock that has a move larger than 15% in the past 90 days” ->Are we talking about :
(1) A single day move of 15+%? OR
(2) Stocks with a quarterly gain of 15+%?
Hi Andreas,
First, thanks for doing the book. Its very refreshing to have some clear presentation in this area (warts and all).
Since we have some degree of correlation across all stock mkts/indexes is there any reason we wouldn’t rank across the entire stock universe rather than a narrow index (perhaps with some mkt cap filter and secondary geographic/sector diversification)? In other words why would we ever choose a narrow versus a much broader pool stocks (eg a deep global index)? If we’re searching for momentum, why artificially limit our search since price is price? Why wouldn’t we include every stock we could possibly buy in our ranking for portfolio inclusion?
Sure, go ahead. The main issue is in how to backtest such an approach. You’ll need to figure out how your total investment universe would have realistically looked ten or twenty years ago. That’s when it gets really convenient to use an index. By using historical index membership, you’ll have a realistic historical universe. If you use a market cap filter or similar, that’s fine, but for backtest you’ll need to find a way to replicate that. To backtest, you’d need to include all delisted tickers, account for mergers and capital events, get hold of historical shares outstanding data to calculate market cap filter etc. It can be done for sure, but it’s not all that easy.
Andreas,
Thanks for replying.
I guess I’m assuming that we’re dealing with common herd behaviour and it’s traits in the induction of trending prices.
If backtesting over the S&P works for a simple momentum strategy then I’m extrapolating that as nominal evidence to all instruments that are traded by humans in markets of sufficient scale. It seems that the only thing we are relying on is that a small but definite proportion of instruments will trend more strongly than others for a sufficient extent to offset the losers. And, that in an upwards trending mkt, there will always be instruments trending more strongly than others.
This seems to just be the definition of a persisting trend within a trending market and that everything else is an attempt at optimization (and that back testing might not be the best guide for that optimization).
My thoughts would not be to necessarily rank and select stocks directly from some global index but to rank all available indexes/mkts for momentum and then rank instruments within those indexes and take positions accordingly (whilst recognizing the additional benefits of reduced correlation that different indexes might offer).
Of course this is just my easy speculation (ie there may well be specific behavioural aspects to market making herds in different markets that backtesting might unearth).
Hi Andreas.
Very interesting book. I may have missed this in your book but do all your simulation opens and closes occur at end of day close prices on the day signals are triggered?
Thank you,
MB