Background and Historical Perspective
Summit Investment Management, Ltd., founded in 1987, is an independent, employee-owned, registered investment advisory firm. We offer actively managed, computer-driven investment portfolios for individual and institutional investors. Since inception of our quantitative strategies in 2002, investors have substantially outperformed the S&P 500 and Russell 1000 benchmark indices.
Summit began testing quantitative/fundamental strategies in late 1992 with the intention of constructing an enhanced equity index portfolio. Our investment strategies were applied to Standard & Poor’s Mid Cap 400 Index. The basic objective was to, once a month, divide the 400 stocks into three groups: overweight the top third, equal-weight the middle third and underweight the remainder. The project lasted about 25 months and it provided some very interesting results. Between 12/11/92 and 1/13/95, the over weighted group (i.e., top one-third) increased 19.02% while the equal-weighted group rose 9.61% and the bottom third group declined 1.17%, compared to the S&P 400, which rose 9.49%. While the top-rated group performed substantially better than the market, we were not able to capitalize on this strategy primarily given the lack of a platform to operate from.
Although we continued to utilize various quantitative strategies to provide a broad list of better quality stocks to cherry pick from, our process was refined and formalized during the summer of 2001. With historical data and better software then available to us, we were able to back test our refined strategies on our select universe of 1000 stocks. The objective was to divide the universe into five parts (quintiles) with the first quintile to represent the top (best) 20% of the universe, and the fifth quintile the bottom 20%. The ranking system was utilized each month and the groups were changed to reflect the new information generated. As can be seen in the below, the results of the test came in well above the market and our expectations.
| BACKTEST RESULTS – 7/31/00 TO 6/30/01 | ||
Total Return |
Difference vs. S&P 500 Index |
|
| Top Quintile | +20.61% |
+35.04% |
| Second Quintile | +9.02% |
+23.45% |
| Third Quintile | -1.61% |
+12.82% |
| Fourth Quintile | +12.21% |
+2.22% |
| Fifth Quintile | -26.26% |
-11.83% |
| S&P 500 (Price Only) | +14.43% |
|
Not only did the quintile groups perform in the exact order as hoped for, the top quintile substantially outperformed the S&P 500, making it an excellent source from which to select investment opportunities. As such, we used the list to harvest candidates for our client portfolios over the next 15 months and kept close watch on how well the top 50, 100 and 200 stocks of our universe performed each month.
Start Date |
End Date |
Top 50 |
Top 100 |
Top 200 |
S&P 500 (Price Only) |
8/3/01 |
9/4/01 |
-1.32% |
-1.62% |
-1.89% |
-6.70% |
9/4/01 |
10/5/01 |
-7.30% |
-8.30% |
-6.53% |
-5.40% |
10/5/01 |
11/2/01 |
-1.62% |
1.23% |
0.51% |
1.50% |
11/2/01 |
12/7/01 |
8.07% |
7.49% |
8.45% |
6.50% |
12/7/01 |
1/4/02 |
7.20% |
6.08% |
3.01% |
1.20% |
1/4/02 |
2/1/02 |
2.10% |
2.15% |
-.05% |
-4.30% |
2/1/02 |
3/1/02 |
2.40% |
3.79% |
-0.18% |
0.90% |
3/1/02 |
4/5/02 |
1.81% |
1.49% |
1.87% |
-0.80% |
4/5/02 |
5/3/02 |
5.96% |
4.80% |
3.18% |
-4.40% |
5/3/02 |
6/7/02 |
-2.37% |
-3.23% |
-3.66% |
-4.30% |
7/5/02 |
8/2/02 |
-14.70% |
-12.40% |
-12.40% |
-12.60% |
8/2/02 |
9/5/02 |
7.30% |
5.70% |
4.30% |
1.70% |
9/5/02 |
10/3/02 |
-10.36% |
-8.80% |
-7.33% |
-6.05% |
10/3/02 |
10/30/02 |
5.80% |
4.49% |
4.86% |
8.16% |
Total Return (15 Months) |
-5.04% |
-5.56% |
-13.23% |
-27.06% |
|
vs. S&P |
+22.02% |
+21.50% |
+13.83% |
The real time paper portfolio results were once again very good. While we were able to select a number of good investments from the top 200, we often commented that we would substantially reduce the risk of our clients’ portfolios if we were somehow able to buy the entire list. Fortunately, the platform (FolioFN) to accomplish this is now operational and in late 2002, we began accepting equity investment funds into portfolios. These portfolios are now refreshed weekly, monthly or quarterly and include either the top 50 or top 100 stocks that are uncovered by our multi-factor selection model. Actual results for a number of our portfolios have been as follows:
Portfolio Returns vs. Benchmarks Thru 13/21/06 |
 |
PAST PERFORMANCE IS NO INDICATION OF FUTURE RESULTS.