From the Volume and Mean Reversion post, a reader sent a suggestion to instead use the ratio of 10 day moving average of the Close times Volume divided by the 63-day moving average of the Close times Volume (CV10/63). I had not tried this before and wanted to see how well it would work.
Overall, I have had very little success integrating volume into any of my strategies. Either volume would have no predictive value or if it did, using it reduced the number of trades too much to be worthwhile. It has been a long while since I have looked into this and I had some new ideas.
A common question I get is whether mean reversion is still working. My response is I am still trading a mean reversion strategy but the edges seem to get smaller. Over the year I have investigated this. I was asked again recently and wanted to investigate again. Here are the results of my 2022 investigation.
Test date range 1/1/2000 to 9/30/2022. I wanted to keep the rules simple. I tested various ways with the 200-day moving average. The reason for this is that some people only trade stocks above the 200, while I like to trade without. In general, this will have more volatility but better returns.
I am drawn to ETF rotation strategies. What likely draws me to them is that in general, these are simple strategies that do not trade that often. My goal with these strategies is to match buy and hold with less drawdown.
What follows is a strategy I have known about for a while and tested but never written about.
This post was corrected on January 18, 2023, after an error was discovered in the code. What is rare, the results with the correct code are better.
From a set of ETFs, select the one to three that have had the best short-term return, best mid-term-return with the least volatility.
In my last blog post, Using Historical Volatility for Parameter Adjustment, I tested using historical volatility to determine trade rules. While reading the July 2022 Technical Analysis of Stocks & Commodities, I came across an article, “Is It Too Volatile To Trade?” by Perry Kaufman. I always like his work so I was interested to see what he had to say. He uses standard deviation from the median historical volatility to decide if a stock is too volatile. He points out that even though returns may be positive during volatile times, it comes with higher risk.
From my own research, I frequently use historical volatility (HV) in my strategies. Most of the time focusing on high HV because these are the stocks that are moving. But when I am trying to tame down drawdowns, I change the focus to low HV stocks. Kaufman’s concept is similar but not something I had tried.
The AllocateSmartly website often has interesting posts. Recently I was reading the article Trending Fast and Slow and thought about other ideas to test. The article is based on research on trading the SPX and depending on the current historical volatility one would either use a 12-month or a 1-month lookback to decide whether to enter or exit the trade. I had tried similar ideas before but not this one.
Given the current rise in inflation, there has been a lot more interest in assets that do well during these times. Gold is one asset that is frequently brought up as an inflation hedge. I have also seen more lately about combining these two into a portfolio.
While talking to a trader, he mentioned an article in the December 2021 issue of Technical Analysis of Stocks & Commodities about Benford’s Law. I had read the same article and was wondering how it could be applied to my trading. Benford’s Law is often used to look for fraud. I am sure I am not committing fraud on myself. As we talked, we wondered whether this could help in selecting which strategy to trade from an optimization?
According to Benford’s Law, in a large set of numbers the significant digit is not evenly distributed but the digit “1” occurs the most frequently and the digit “9” the least. The theoretical distribution is shown in this chart.
In general, for a data set to follow Benford’s Law, the data set should have the following properties
For more information about Benford’s Law, these are two good articles. What is Benford’s Law and why is it important for data science? and Wikipedia’s article Benford’s law.
Using the percentage daily return on stock data, here is how the data properties fit with what Benford’s Law wants in a data set
The data are not a perfect fit, but we will try and see what we get.
From 2007 to 2021, here are the results of the significant digit count on the daily percentage return on the SPY
Visually the SPY appears to follow Benford’s Law. There are several statistical methods for determining if it does follow. I will be using the Chi-Square Statistic. What we need to know about this value is the lower the better. If there was a perfect match between the data set and the theoretical distribution, the value would be zero. Typically, a value under 15.5 is used to determine if the data does follow Benford’s Law. The Chi-Square Statistic for the SPY data is 17.9. Close to the cutoff. I guess perhaps the SPY is slightly manipulated.
Next, I was curious about how the individual stocks of the S&P500 behaved. I used the current S&P 500 stocks and data from 2007 to 2021.
Values for the Chi-Square range from a low of 5 for MTB to a high of 102 for AWK. The average for all was 44.4. Only seven stocks had values under 15.5. Either one of two things is going on here. One, these stocks are highly manipulated. Or two, the data is not a good fit to apply Benford’s Law. I am going with number two because I am an optimist and believe the markets are only slightly manipulated.
Now to what I wanted to know. Is there any predictive value here for selecting a strategy from an optimization run?
I ran an optimization (432 runs) on a momentum strategy that I trade from 2007 to 2016. I took the top 1/3 of the runs (145 runs) based on CAR and computed the Chi-Square Statistic for each based on the daily % return of the strategy. From the Chi-Square Statistic, I divided these into terciles.
Side note: I thought the Chi-Square Statistic was big for the S&P 500 stocks. In my optimization run, the range was from 210 to 2143. Not even close to following Benford’s Law!
Then using an out-of-sample optimization, I computed the CAR tercile that each of these in-sample runs fell into. My hypothesis was that failure to follow Benford’s Law implied a higher potential for curve fitting, which then implies that those in-sample runs in the top tercile for CAR and bottom tercile for Chi-Square Statistic would again end up in the top tercile for CAR in the out-of-sample.
The numbers in the table represent how many runs fell into that cell, with a blue background highlighting the highest value in that row The first row represents 48 runs from the in-sample that fell into the top tercile by CAR and then the top tercile by Chi-Square Statistic for that top tercile. The blue 30 then means that 30 of those 48 runs end up in the top tercile by CAR in the out-of-sample.
At first, I was excited by the very clear diagonal line and the fact that the other cells followed what I expected. But then I realized this was the opposite of what I thought. I expected the lowest tercile by Chi-Square Statistic to be most likely to stay in the top CAR tercile. I started making up stories to explain this, but then I realized I should test this on a different strategy.
Following the same process that I used for the momentum strategy, here are the results of a mean reversion strategy.
Now, it is random. No pattern at all. Too bad.
I want to thank Matt Radtke for helping proof this article. The first time I sent it to him, he caught a huge logical mistake that I had made. It would have been embarrassing to publish an article with such a bad mistake. That is also the reason it has been so long since my last post. I had to redo a lot of the work I had done.
As often happens in research, an interesting idea leads nowhere. Given that the data do not follow the requirements for applying Benford’s Law, I am not too surprised by the results. This was a good lesson for me on not stopping when I got the first good results. Even had the results panned out, I am not sure I would have used them. No story that I could make about why this should work made me comfortable. And we must be comfortable with the reasons we are trading our strategies.
Backtesting platform used: AmiBroker. Data provider: Norgate Data (referral link)
I’ve been writing this blog for nine years now. Sometimes I am amazed about topics I have not covered and this is one of them. When developing a new strategy, these are the indicators I likely test: RSI, Historical Volatility and Internal Bar Strength (IBS). I had a reader send me an email pointing me to research done on IBS. I thought let me send him my blog post on this. After searching my records and site, I could not find one. That made it easy to decide what my next post would be about.
I’ve been seeing lots of seasonality type charts on the S&P500 where they take the average return for each day of the year and then create a return curve for the year. The chart often ‘shows’ the sell in May and buy in November flatness of the returns. And then the holiday end of the year run up.
Steven, my trading buddy, sent me yet another chart and I noticed something I had not seen before. A beginning of the year downtrend from January to mid-March. This got me thinking. How much does the start of the data set impact these charts?