8. LTM EPS Yield

Description of Factor:

The LTM EPS Yield is the inverted LTM P/E ratio, LTM being the last twelve months. LTM is chosen since it reflects the most recent financial information available over a yearly basis without having to constitute the end of a company’s fiscal year. The rationale behind this variable is the expectation that companies with a high yield (corresponding to a low P/E), will outperform the market since those stocks are “cheap” on a relative basis. From an investment strategy standpoint, this would imply value focused strategy.

Analysis

Equal Weighted: As can be seen from the graph below, the quintiles nearly follow a perfect step distribution with quintile 1 having the highest annualized average return of 36.39% and quintile 5 the second lowest annualized average return with 12.39% (quintile 4 is slightly lower at 11.53%). The standard deviation for quintile 1 is 16.59% while it numbers 19.29% for quintile 5. The average excess return of quintile 1 is 14.75% and (5.52)% for quintile 5. The beta for quintile 1 is 1.07 with an R² of 71%, while the beta for quintile 5 is 1.16 with an R² of 62%.

Using LTM EPS yield as a screening factor works well when executing a long-short strategy. Looking at the in-sample and out-of-sample data shows us that quintile 1 had the highest annualized average return for all but one year (1999), whereas the performance of quintile 5 was slightly more volatile, having the worst return of all the portfolios in only six years. During five years, quintile 5 was the second worst performing quintile. It was the middle performing quintile for two years. It is important to that in 1999, we would have lost 68% if adhering to our long-short strategy. It is astounding that the cumulative return over the sixteen years for quintile 1 is a whopping 24,685%. The monthly turnover for quintile 1 is 13.76%, and 8.16% for quintile 5.

Value Weighted: The value weighted portfolio’s performance mimics that of the equal weighted portfolio. Quintile 1 outperforms all quintiles and quintile 5 outperforms quintile 4 by a very slight margin. Quintile 1 yields an annualized average of 33.71% and quintile 5 returns 13.99%. The standard deviation for quintile 1 is 16.47% while it numbers 16.12% for quintile 5. The average excess return of quintile 1 is 12.65% and (4.27)% for quintile 5. The beta for quintile 1 is 1.18, while beta for quintile 5 is 1.11.

The performance of the value weighted portfolio is not as compelling as the equal weighted portfolio. Out of the sixteen periods, quintile 1 generates the highest return for twelve years, whereas quintile 5 generates the lowest return in only seven years.

Conclusion:

Based on the analysis above, it is clear that the equal weighted portfolio produces better results. Although the LTM EPS yield, on an equal-weighted basis, produces the highest return portfolio well, the one year it fails would have been disastrous. This year was 1999, when investment fundamentals were not closely watched by investors and low EPS yield stocks (high P/E) did very well. Although quintile 5 performs poorly each year, it fluctuates frequently between the worst performing and the second worst performing portfolio. In two periods it is the third best performing portfolio. Therefore, the implementation of a long-short strategy based solely on the LTM EPS Yield factor is not advisable, but use of the factor may be attractive in the connection with others.