Lessons from David Dreman (Part 2)
Posted by Mark on May 31, 2016 at 06:30 | Last modified: April 21, 2016 05:43David Dreman uses a contrarian investing approach in an attempt to profit from irrational heuristic thinking. Today I want to discuss Dreman’s suggestions to protect yourself from such harmful cognitive biases.
With regard to the representativeness heuristic, Dreman suggests focusing on relevant factors that may result in an entirely different investment result. Do not be blinded by similarity. Sometimes only a slight similarity is necessary to significantly resonate with us especially if a big profit/loss was realized in the previous instance.
Dreman cautions against being influenced by short-term returns of a money manager, analyst, or any kind of financial adviser. The better the numbers, the more alluring the results but this is based on an insufficiently small sample size. Look for stronger supporting evidence like a longer term track record.
Dreman suggests being doubly cautious of short-term returns when they deviate significantly from long-term averages. This applies to performance of money managers, analysts, and advisers as well as to individual stocks or funds. Mean-reversion is the rule, not the exception.
Dreman recommends patience with new investment strategies rather than expectation of quick success. Investors are more likely to adopt strategies that have recently done well, which implicates mean-reversion as an immediate headwind. One way to offset this would be to enter a strategy when it seems somewhat out-of-favor. Such a strategy has already mean-reverted or is broken altogether, in which case future outperformance will no longer result. Only in retrospect can these two possibilities be differentiated.
These are four ways to prevent biases from affecting your investing decisions.
Dreman suggests purchasing out-of-favor stocks (e.g. low price-to-earnings ratios) as one way to profit from the biases of others. While this idea sounds good, as with most other strategies I would be interested to see a large backtest to better understand context.