Factors used in our model - Fuqua School of Business



Factors used in our model

For each of the factors we use the sort algorithm mentioned before, then we calculated a series of ratios for each fractile to determine which of the factors were more appropriate to outperform the benchmark. As we mentioned before, we are using geometric average of returns as well as the value weighted portfolioFollowing there is an example of the information we calculated for each factor.

Top Fractile

| |Average excess return |Annual Volatility |Hitting ratio (% |

| |(Basis points) | |outperformance) |

|Earnings Yield |627 |37.81 |54% |

|Dividend Yield |-112 |40.79 |52% |

|Book to Value Yield |382 |37.59 |53% |

|Cash Flow Yield |874 |46.67 |64% |

|IBES 1 month revision ratio |-384 |57.49 |53% |

|Prospective Price Earnings 12 Month with Last Fiscal Year|613 |40.11 |57% |

|Prospective Earnings Growth Yield 12 Month with Last |275 |46.74 |52% |

|Fiscal Year | | | |

|Revenue Growth Rate |97 |43.60 |48% |

|Earnings Growth Rate |135 |43.66 |49% |

|Prospective Price Earnings 12 month forward |12 |37.67 |47% |

Bottom Fractile

| |Average excess return |Annual Volatility |Hitting ratio (% |

| |(Basis points) | |outperformance) |

|Earnings Yield |-458 |42.01 |37% |

|Dividend Yield |-520 |38.03 |45% |

|Book to Value Yield |-757 |37.73 |39% |

|Cash Flow Yield |-54 |49.55 |50% |

|IBES 1 month revision ratio |-377 |53.48 |40% |

|Prospective Price Earnings 12 Month with Last Fiscal Year|-637 |40.36 |42% |

|Prospective Earnings Growth Yield 12 Month with Last |-330 |36.17 |45% |

|Fiscal Year | | | |

|Revenue Growth Rate |-418 |37.51 |43% |

|Earnings Growth Rate |-263 |37.95 |52% |

|Prospective Price Earnings 12 month forward |-863 |46.41 |43% |

| | |

| | |

| | |

|Earnings Yield (EY) | |

| | |

|The top fractile portfolio, based on this factor sorting, outperformed the | |

|index under a buy and hold strategy by 627 basis points with a lower | |

|volatility (37.81 vs. 42.01). However, this factor is less accurate | |

|predicting outperformers when the market went up (48% hitting ratio when | |

|market went up vs. 57% when market down). The low 48% hitting ratio when | |

|the market goes up suggests that this factor does not work properly under | |

|that situation. The bottom portfolio clearly under-performed the benchmark,| |

|giving important information of what stock would potentially | |

|underperformed. | |

| | |

|Even though the top portfolio outperformed the benchmark for the entire | |

|period, there were some years where it did not outpaced the market (1993, | |

|1995 and 1998) – 3 years out of 10. The bottom portfolio outperformed the | |

|market in the first five month of 1998, it would be interesting to further | |

|analyze why. | |

| | |

|The top fractile portfolio has a beta of 0.9, while the bottom has a beta | |

|of 1.03. | |

| | |

| | |

| | |

|Dividend Yield (DY) | |

| | |

|Based on this factor sorting, none of the fractiles outperformed the index | |

|under a buy and hold strategy. Although this factor has an average hitting| |

|ratio higher than 50% for the entire period, it is basically the result of | |

|an extremely good year in 1992. Both portfolios underperformed the | |

|benchmark by more than 100 basis points for the entire period. | |

| | |

|The graphs of yearly returns and the index returns clearly shows how the | |

|difference in returns between the top fractile and the market is very | |

|small, but usually the top portfolio underperformed the benchmark. | |

| | |

|An interesting fact is that both portfolios has lower than one beta, but | |

|their returns were lower than that of the market. Even their | |

|mean-variance adjustment is lower than that of the market. | |

| | |

| | |

| | |

|Book to Value Yield (BV) | |

| | |

|The top fractile of this factor outperformed the benchmark by 382 basis | |

|points while the bottom fractile underperformed it by 757 basis points | |

|under the strategy buy and hold for the entire period, both with lower | |

|volatility. Although the hitting ratio is not that high when the market | |

|goes up (50%), the excess returns are significantly higher between | |

|1992-1994 (when the Korean market went constantly up), but after 1995 the | |

|excess return was not than high (see yearly return and index graphs). | |

| | |

|The bottom portfolio clearly underperformed the benchmark for 8 of out 10 | |

|years. This can be an insightful information for the “sell” list. | |

| | |

| | |

| | |

|Cash Flow Yield (CF) | |

| | |

|The top fractile resulting from this factor had the most impressive | |

|performance . It outperformed the market by 874 basis points with an | |

|average hitting ratio of 64%. However, the data available for it is from | |

|1993, which does not incorporate the first “cycle” of the Korean index. | |

|But the results are promising, the top fractile had positive excess returns| |

|5 out of the 6 years. And the year that it did not outperformed was 1998, | |

|which is just an annualized return for the first five months of 1998. | |

|Another important element is that this top fractile also had consistent | |

|higher than 50% hitting ratio for both, up and down markets. | |

| | |

|The volatility for the top fractile is somewhat higher than that of the | |

|market, although it has a beta of 0.99. | |

| | |

| | |

| | |

|Prospective Price Earning Yield 12 months with Last Fiscal Year | |

|(PPE_12MWITHFY1) | |

| | |

|The top fractile in this case has an excess return of 612 basis point, | |

|while the bottom fractile has an excess return of –637 basis points for a | |

|buy and hold strategy for the entire period. The top fractile has a good | |

|average hitting rate of 57%. However, breaking it down by upward and | |

|downward market trend, its is clear than this fractile does not do a good | |

|job predicting positive excess return when the market goes up (48% hitting | |

|ratio). Similarly, although it seems to do an extraordinarily good job | |

|outperforming the market when it goes down (63% hitting ratio), in the year| |

|by year return graphs can be seen that it primarily a consequence of one | |

|year, 1992. | |

| | |

| | |

| | |

|Return Growth Rate (RGR) | |

| | |

|Even though the top fractile of this factor has 97 basis point of excess | |

|return for the entire period, its month by moth hitting ratio is very poor | |

|(48%) and it is not very stable, as the index graph shows. | |

| | |

|Its top fractile is riskier than the market as its beta is 1.04 and its | |

|volatility higher than the market. | |

| | |

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download