Book Reviews, Trading

Book Review of The Janus Factor: Trend Follower’s Guide to Market Dialectics by Gary Anderson

The full title of this book is The Janus Factor: Trend Follower’s Guide to Market Dialectics by Gary Edwin Anderson (2012).

This book presents a trading strategy based on relative strength and relative momentum. It classifies the market into two regimes: trending, and contrarian. In a trending regime, strong stocks get bought, weak stocks get sold, so the “spread” in the performance between strong and weak stocks increases. In a contrarian regime, strong stocks get sold, weak stocks get bought, so the “spread” tightens. By measuring the relative strength spread, you determine what regime you are in.

Gary then creates another measure: the Direction of Momentum (DOM) to determine, during periods when it is a trending regime, whether the trend is up or down. Then the strategy, at a high level, is to go long high relative momentum stocks when the trend is up, go short low relative momentum stocks when the trend is down, and stay in cash during contrarian periods.

There are differences in how the different measures are being calculated. For example, relative strength is calculated based on two components, which are obtained based on how a stock performs during periods when the benchmark returns are above average, and during periods when the benchmark returns are below average.

This is a pretty unique piece of work, and it is good stuff that Gary kept building layer upon layer on his fundamental concepts until he has a trading strategy.

Trends Exist

  • Tom Basso: If there were no trends, you could expect a fairly random distribution of price changes. Yet if you look at the distribution of price changes over time in almost any market, you’ll see a very long tail in the direction of large price changes. This is because there are abnormally large price changes that you’d never expect to see by chance over a given period of time.

Only Trends Offer Profit

  • It is the trend, and only the trend, that offers profit. Overbought/oversold indicators, trading setups, and patterns we read into charts might help clarify, or not, might embolden the trader to take a position, or not. But there is no device or trading trick that makes the trader money. Only the trend can do that.

Measure Fund Managers Using Offensive and Defensive Returns

  1. Calculate benchmark returns
    • Using 5-years (20 quarters) of data, calculate the median quarterly performance across all managers for each quarter (“benchmark returns”).
    • Calculate the benchmark median quarterly return across the 20 quarterly returns.
  2. Categorize each quarter as ‘offensive’ or ‘defensive’
    • Benchmark quarterly returns that are greater than or equal to the “benchmark median quarterly return over the 20 quarterly returns” are labelled ‘offensive returns’, else they are labelled ‘defensive returns’.
  3. Categorize a manager’s returns using the same labels each quarter
    • If the benchmark’s quarterly return falls under ‘Offensive,’ then the manager’s return is also categorized under ‘Offensive’; else the return is categorized as ‘Defensive’.
  4. Calculate manager’s scores
    • Offensive Score = Sum of the manager’s quarterly offensive returns over the 5 years / Sum of the benchmark’s offensive returns for the same quarters * 100
    • Defensive Score = Sum of the manager’s quarterly defensive returns over the 5 years / Sum of the benchmark’s defensive returns for the same quarters * 100
    • Defensive score below 100 indicates better-than-benchmark defense.
    • By definition, benchmark is at 100 for both offense and defense.
  5. Draw an axis of offensive score vs. defensive score with (100, 100) at the center.
    • When you multiply benchmark returns by a factor, you can scale up or down the volatility. These different combinations differ in volatility but are equivalent in relative performance. So if you draw a +45-degree line linking across points (0, 0) to (100, 100), you get the Benchmark Equivalence Line (BEL).
    • Managers NW of the BEL outperformed the benchmark. SE of the BEL underperformed.
    • SW managers are less volatile than NE managers.

Application to Stocks

The same matrix can be applied to stock returns

  • Use stock daily returns over a 6-month look back period (analogous to 20 quarters over 5 years).
  • Relative strength (RS) of a stock
    • = distance to the BEL line = (offensive score – defensive score) / sqrt(2).
    • RS is assigned positive if NW of BEL, negative if SE of BEL.
  • Relative strength spread (RSS)
    • = average RS of the strongest 10% of stocks – average RS of the weakest 10% of stocks.
  • Performance spread (PS)
    • Calculate the cumulative forward performance of RS leaders by identifying the strongest 10% of stocks each day, get their average return over the next day, then repeat for each day moving forward. Do the same for RS laggards.
    • PS =Cumulative forward performance of strongest 10% – cumulative forward performance of weakest 10%
    • PS reveals whether RS strategies are working

Positive vs. Negative Feedback

  • Positive feedback (e.g. trend following)
    • Price rises >> Traders buy
    • Price falls >> Traders sell
    • Relative strength of issues increases (absolute value) and causes expansion: issues NW of the BEL (leaders) go more NW, issues SE of the BEL (laggards) go more SE.
    • Both RSS and PS increases
  • Negative feedback (e.g. contrarian)
    • Price rises >> Traders sell
    • Price falls >> Traders buy
    • Relative strength of issues drop and causes contraction: issues move toward the BEL.
    • Both RSS and PS decreases

Why Jesse Livermore Lost Money From 1911 – 1914

  • After years of extended trends on both directions, prices on the NYSE settled into a range. For a trend follower like Livermore, truncated trends and frequent reversals made trading difficult, if not treacherous.
  • Livermore’s rule: In a narrow market, when prices are not getting anywhere to speak of but move within a narrow range, there is no sense in trying to anticipate what the next big movement is going to be — up or down.
  • PS (75 stocks, monthly data, 7-month look-back, 8% strongest and weakest) fell persistently from 1911 to 1914

Performance Spread Declines Within the Same Industry

  • When the benchmark are stocks in the same industry, the performance spread constantly declines because negative-feedback traders persist in turning outliers near both perimeters of the universe back toward the BEL.
  • Stocks that are significantly stronger than the group average are viewed as overpriced and are offered, while laggards receive a bid. In effect, traders act like sheep dogs.

Performance Spread Falls During Market Tops and Bottoms

  • The initial phase of a bull market always begins as a correction (to the downtrend), driven by negative feedback and signaled by a falling performance spread. However, the initial, contrarian-driven phase of a bull market is indistinguishable from a bear market correction.
  • Contrarian herding is seen not only near bear market bottoms, but as well around bull market tops, though not with the same dependable regularity. However, a contrarian collapse is indistinguishable from other bull market corrections.

Weakest Stocks in a Bear Market are the Best Longs 

  • The territory around bear market lows is prime hunting ground for contrarians seeking bargains. The weakest, most oversold stocks then receive the strongest bids and, as a result, recover faster than stocks that held up relatively well during the preceding decline.

Direction of Momentum (DOM) Measures the Trend Direction

  • Direction of Momentum (DOM) = Cumulative of market’s price changes on days when RSS expands.
  • Direction of Entropy (DOE) = Cumulative price changes on days when RSS contracts.
  • Relative Momentum (RM) = Conventional relative strength calculation but substituting DOM for price.

RM Leaders and Laggards Are Better Bets Than RS Leaders and Laggards

  • We can calculate the RS and RM of industry groups, and compare that to the average performance of the entire universe of groups.
  • RS / RM leaders are groups with RS / RM in the top 8%. To determine the RM leaders / laggards, use six-month sums of daily changes in group DOMs.
  • The forward performance of RM leaders outperformed RS leaders which outperformed the group average.
  • The forward performance of RM laggards underperformed RS laggards which underperformed the group average.

Long-Only DOM Strategy

  • Determine a Critical Moving Average (CMA) length
    • Try EMAs of lengths from 29 to 60 days.
    • When the EMA of the DOM for the average group is rising, take a long position. If EMA is falling, take a short position. Do this every day and calculate the hypothetical returns.
    • The EMA length that gives the best profits is deemed the CMA.
  • DOM-only strategy
    • Go long RM leaders when CMA of DOM is rising, exit when CMA of DOM is declining.
    • Do this every day, and recalibrate the CMA every day.
    • One flaw is when there is a contrarian collapse (e.g. contrarian activity at market tops), the DOM does not warn of the threat by declining, it continues to indicate bullish momentum.
  • Defensive strategy
    • Go long RM leaders when CMA of DOM + CMA of RM PS are both rising, else stay in cash.
    • To find length of CMA of RM PS, test EMAs ranging from 14.5 to 30 days.
    • This attenuates corrections but there is a long-term performance impact.

Short Strategy

  • Go short RM laggards when DOM is falling and RM PS is rising.
  • Go to cash when DOM is falling and RM PS is falling.

Long-Short Strategy

  • Long when DOM is rising
  • Short when DOM is falling, RM spread is rising.
  • Cash when DOM is falling, RM spread is falling.



No comments yet.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Copyright © 2005-2014 All Rights Reserved.

Enter your email address to follow this blog and receive notifications of new posts by email.

Blog Stats

  • 398,148 hits
%d bloggers like this: