The Moving Average Research King: Valeriy Zakamulin

/The Moving Average Research King: Valeriy Zakamulin

The Moving Average Research King: Valeriy Zakamulin

By | 2017-08-18T16:55:25+00:00 April 22nd, 2016|Tactical Asset Allocation Research|5 Comments
Print Friendly, PDF & Email
(Last Updated On: August 18, 2017)

Some weekend reading for trend-followers who want to question their beliefs.

Valeriy Zakamulin is an animal when it comes to generating research on moving averages. We’ve done a lot of the same work, but we’re too lazy to tabulate the results in an academic paper format.

king of ma

The king of moving average research.


Check these papers out:


Revisiting the Profitability of Market Timing with Moving Averages

In a recent empirical study by Glabadanidis (“Market Timing With Moving Averages”  (2015), International Review of Finance, Volume 15, Number 13, Pages 387-425; the paper is also available on the SSRN and has been downloaded more than 7,500 times) the author reports striking evidence of extraordinary good performance of the moving average trading strategy. In this paper we demonstrate that “too good to be true” reported performance of the moving average strategy is due to simulating the trading with look-ahead bias. We perform the simulations without look-ahead bias and report the true performance of the moving average strategy. We find that at best the performance of the moving average strategy is only marginally better than that of the corresponding buy-and-hold strategy. In statistical terms, the performance of the moving average strategy is indistinguishable from the performance of the buy-and-hold strategy. This paper is supplied with R code that allows every interested reader to reproduce the reported results.

The current version of the paper on the SSRN                                    Download the R code and data

A Comprehensive Look at the Empirical Performance of Moving Average Trading Strategies

by Valeriy Zakamulin, working paper
Despite the enormous current interest in market timing and a series of publications in academic journals, there is still lack of comprehensive research on the evaluation of the profitability of trading rules using methods that are free from the data-snooping bias. In this paper we utilize the longest historical dataset that spans 155 years and extend previous studies on the performance of moving average trading rules in a number of important ways. Among other things, we investigate whether overweighting the recent prices improves the performance of timing rules; whether there is a single optimal lookback period in each trading rule; and how accurately the trading rules identify the bullish and bearish stock market trends. In our study we, for the first time, use both the rolling- and expanding-window estimation scheme in the out-of-sample tests; study the performance of trading rules across bull and bear markets; and perform numerous robustness checks and tests for regime shifts in the stock market dynamics. Our main results can be summarized as follows: There is strong evidence that the stock market dynamics are changing over time. We find no statistically significant evidence that market timing strategies outperformed the market in the second half of our sample. Neither the shape of the weighting function nor the type of the out-of-sample estimation scheme allows a trader to improve the performance of timing rules. All market timing rules generate many false signals during both bullish and bearish stock market trends, yet these rules tend to outperform the market in bear states.

The current version of the paper on the SSRN

Market Timing With a Robust Moving Average

by Valeriy Zakamulin, working paper

In this paper we entertain a method of finding the most robust moving average weighting scheme to use for the purpose of timing the market. Robustness of a weighting scheme is defined its ability to generate sustainable performance under all possible market scenarios regardless of the size of the averaging window. The method is illustrated using the long-run historical data on the Standard and Poor’s Composite stock price index. We find the most robust moving average weighting scheme, demonstrates its advantages, and discuss its practical implementation.

The current version of the paper on the SSRN

Anatomy of Market Timing with Moving Averages​

by Valeriy Zakamulin, working paper
The underlying concept behind the technical trading indicators based on moving averages of prices has remained unaltered for more than half of a century. The development in this field has consisted in proposing new ad-hoc rules and using more elaborate types of moving averages in the existing rules, without any deeper analysis of commonalities and differences between miscellaneous choices for trading rules and moving averages. In this paper we uncover the anatomy of market timing rules with moving averages. Our analysis offers a new and very insightful reinterpretation of the existing rules and demonstrates that the computation of every trading indicator can equivalently be interpreted as the computation of the weighted moving average of price changes. This knowledge enables a trader to clearly understand the response characteristics of trading indicators and simplify dramatically the search for the best trading rule. As a straightforward application of the useful knowledge revealed by our analysis, in this paper we also entertain a method of finding the most robust moving average weighting scheme. The method is illustrated using the long-run historical data on the Standard and Poor’s Composite stock price index. We find the most robust moving average weighting scheme and demonstrates its advantages.

  • The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Alpha Architect, its affiliates or its employees. Our full disclosures are available here. Definitions of common statistics used in our analysis are available here (towards the bottom).
  • Join thousands of other readers and subscribe to our blog.
  • This site provides NO information on our value ETFs or our momentum ETFs. Please refer to this site.

About the Author:

After serving as a Captain in the United States Marine Corps, Dr. Gray earned a PhD, and worked as a finance professor at Drexel University. Dr. Gray’s interest in bridging the research gap between academia and industry led him to found Alpha Architect, an asset management that delivers affordable active exposures for tax-sensitive investors. Dr. Gray has published four books and a number of academic articles. Wes is a regular contributor to multiple industry outlets, to include the following: Wall Street Journal, Forbes,, and the CFA Institute. Dr. Gray earned an MBA and a PhD in finance from the University of Chicago and graduated magna cum laude with a BS from The Wharton School of the University of Pennsylvania.
  • Mark

    Hi Wes,

    Interesting post. I took a brief look at these papers and it seems that employing market timing rules still leads to improved risk adjusted returns, mostly because it gets an investor out of bear markets. So I’m assuming this research reinforces the recommendations you make in the DIY strategy, is that correct?

    Also, have you considered more advanced market timing methods like GARCH models which factor in volatility clustering, do they add value to the simple MA approaches discussed here?


  • Yeah, trend rules are really meant to provide tail-risk protection. They don’t necessarily eliminate volatility (ie std dev) and tracking error relative to the market is going to shoot up (because you’re out sometimes). Tracking error shouldn’t matter for a goals-based investor, but for a lot of folks it does.

    Yeah, we looked at volatility a few years ago and burned way too many brain cells on the topic. Conclusion: Volatility rules that work are trend-following in another clothing. Why? Well, when the market blows up, volatility blows up, and the rules tell you to get out of the market. What do trend rules do? When the market blows up, trend rules break, and you get out of the market…so I don’t think they are any better/worse than trend…i think they are the same/similar.

  • dph

    Even a simple 200 day MA seems to work as well as more complicated strategies.

  • Mark

    Hi Wes,

    In the 2nd paper you linked to ‘ A Comprehensive Look at the Empirical Performance of Moving Average Trading Strategies’ the author distinguishes two periods of market activity; period:1, 1870-1942, and period:2, 1942-2014. He shows that MA trading strategies have value in period:1 (MOM rule has Sharpe ratio of .46 vs. .25 for market) but not so much in period:2 (MOM rule has Sharpe of .63 vs. .56 for market) , and argues that period:2 is the more relevant indicator of the future returns of MA strategies. I think the last point is questionable, it seems to me there are reasonable arguments to be made that the future will look more like the first period. Namely lower return with significant periods of stock market volatility. My take on this is that if the market behaves like period:2 then you pay a small price in overall performance for crash protection, but get similar risk adjusted returns and presumably a better Calmar ratio. If the market behaves like period:1 then the crash protection inherent in market timing is well worth it.

    Also, from a portfolio management perspective does the crash protection afforded by an MA rule warrant a higher allocation to equities over bonds? If so, this seems like another good reason to use an MA rule in today’s low interest rate environment.

  • All reasonable ideas.
    I think it is a reasonable hypothesis that a downside-protected heavy equity + light bond portfolio will do better than a static 60/40 stock/bond portfolio in an environment with lower expected returns and high volatility. I’m a believer…but who knows what the market Gods will thrust upon us.