Back To Comparative Analysis
		HAUGEN SYSTEMS ALTERNATIVE SECTOR PORTFOLIO BACKTESTS
	In this test we started with the Greenblatt strategy of "Return on Assets" & "Earning Yield" (see the first sheet									Backtest Parameters:
	for the differences in our factors), and constructed portfolios based on a population of the largest 1,000									Region		U.S.
	stocks.  Fifty stocks were added at the beginning, four stocks were rebalanced each month, and the monthly returns									Population		Top 1,000
	were linked together to compute an annualized return.  Then 11 more tests were run, each starting on a									# of stocks		50
	different month in 1996, but otherwise employing the same rules.  Finally, the 12 test's annualized returns were									First Date		1996
	averaged and shown below.									Last Date		12/31/05
	So the first column shows average portfolio returns for the 12 tests done on Return on Assets & Earnings Yield.									Sector Constrain		No
	The next columns show how the portfolios did without stocks from certain sectors.  For example, the 2nd column									Rebalance periodicity		Monthly
	has all the same rules as the first, except stocks from the finance and utility sectors were excluded from the tests.									Number to rebalance		4
	Greenblatt						Greenblatt	Haugen	S&P
	Return on Assets	Minus	Minus	Minus	Minus	Minus	Returns	Model	500
	& Earnings Yld	Finance	Durable	Energy	Construct.	Trans.	(from the		Index
	All	Util	Non-Drbl	Manufact.	Bus. Equip	Bus. Srvc	book)
	Sectors
	---------------------A  V  E  R  A  G  E    P  O  R  T  F  O  L  I  O    T  O  T  A  L    R  E  T  U  R  N-----------------------
1996	30.96%	31.14%	45.42%	34.77%	22.43%	31.03%	37.40%	48.06%	22.96%
1997	11.13%	9.54%	11.15%	15.26%	17.82%	11.66%	41.00%	38.83%	33.36%
1998	8.91%	13.19%	11.25%	9.90%	-0.51%	11.16%	32.60%	29.61%	28.58%
1999	12.35%	8.80%	18.34%	12.40%	4.08%	12.74%	14.40%	42.17%	21.04%
2000	17.28%	14.88%	3.05%	13.48%	23.41%	17.45%	12.80%	-3.53%	-9.11%
2001	7.61%	8.64%	3.80%	7.99%	11.24%	8.92%	38.20%	1.06%	-11.89%
2002	6.50%	7.63%	9.99%	6.69%	2.62%	6.44%	-25.30%	3.78%	-22.10%
2003	31.13%	31.89%	36.10%	29.15%	29.13%	31.79%	50.50%	33.92%	28.68%
2004	23.29%	22.28%	20.47%	20.56%	18.80%	23.38%	27.60%	26.69%	10.88%
2005*	20.43%	18.22%	20.76%	12.16%	19.86%	19.82%	N/A	32.70%	4.91%
Averge Linked Annual Ret.	16.64%	16.31%	17.36%	15.92%	14.49%	17.14%	18.67%	24.05%	9.07%
Average Annual Return	16.96%	16.62%	18.03%	16.24%	14.89%	17.44%	25.47%	25.33%	10.73%
Longitudinal Std.Dev.	9.22%	9.12%	13.64%	9.24%	9.99%	8.94%	22.61%	18.31%	0.195102
T-Stat	5.52	5.46	3.97	5.27	4.47	5.85	3.19	4.15	1.65
Probability Total Ret < 0	2.10%	2.34%	10.21%	3.23%	7.31%	0.63%	13.97%	9.11%	28.43%
	---------------------A  V  E  R  A  G  E    P  O  R  T  F  O  L  I  O    E  X  C  E  S  S    R  E  T  U  R  N-----------------------
1996	8.00%	8.18%	22.46%	11.81%	-0.53%	8.07%	14.44%	25.10%	0.00%
1997	-22.24%	-23.82%	-22.21%	-18.10%	-15.54%	-21.70%	7.64%	5.46%	0.00%
1998	-19.67%	-15.39%	-17.33%	-18.68%	-29.09%	-17.41%	4.02%	1.03%	0.00%
1999	-8.69%	-12.24%	-2.71%	-8.64%	-16.96%	-8.30%	-6.64%	21.13%	0.00%
2000	26.39%	23.98%	12.16%	22.59%	32.51%	26.56%	21.91%	5.58%	0.00%
2001	19.50%	20.52%	15.68%	19.87%	23.12%	20.81%	50.09%	12.94%	0.00%
2002	28.60%	29.73%	32.09%	28.79%	24.72%	28.54%	-3.20%	25.88%	0.00%
2003	2.44%	3.21%	7.41%	0.46%	0.45%	3.10%	21.82%	5.24%	0.00%
2004	12.42%	11.40%	9.59%	9.68%	7.92%	12.50%	16.72%	15.82%	0.00%
2005*	15.52%	13.31%	15.85%	7.26%	14.95%	14.91%	N/A	27.79%	0.00%
Average Annual Return	6.23%	5.89%	7.30%	5.50%	4.16%	6.71%	14.73%	14.60%	0.00%
Longitudinal Std.Dev.	18.04%	17.86%	16.98%	16.56%	20.26%	17.64%	16.94%	9.97%
Information Ratio	0.35	0.33	0.43	0.33	0.21	0.38	0.87	1.46
T-Stat	1.04	0.99	1.29	1.00	0.62	1.14	2.61	4.39
Probability Excess Ret < 0	35.28%	35.85%	32.31%	35.75%	40.70%	34.02%	19.69%	7.74%
	Each test's results were computed as follows:
	1. First the test that began in January had the monthly returns linked together to get an annual total return for 1997.  Linking is the
	geometric average calculated by adding 1 to each monthly return, multiplying the numbers together, and subracting 1.
	2. Then 1998 thru 2002 had the total returns calculated in a similar way.
	3. Then the February portfolio's annual returns were calculated the same way, then March, and so on resulting in a table
	with 6 by 12 annual returns
	4. Then 1997's average annual return was calculated by taking the arithmetic average of all 12 portfolio's 1997 results
	5. The rest of the year's average annual returns were then calculated.
	The Average Linked Annual Return was then computed by linking these average annual returns together.  Like with the monthly
	returns, above, we added 1 to each annual return, and multiplied the numbers together.  Then, that product was taken to the
	12 / 72nd power before subtracting by one.  The numbers from Greenblatt's books were already geometric averages, so the
	average linked annual return could be calculated the same way as in the rest of the columns.
	The Average Annual Return is the arithmetic average of each test's (eg Jan96, Jan97, Feb96, etc) montly total return.
	The longitudinal standard deviation is calculated by taking a standard, non-biased standard deviation of the average annual returns
	The information ratio is the average excess return divided by the standard deviation of the excess returns
	The t-stat is the average annual return divided by the standard deviation divided by the square root of the number of years in the study.
	The probability is calculated by first finding the range of returns in the bell curve (eg, 95% of the returns will be within 2 standard
	deviations) and then calculating the ratio of this range and dividing by 2.
Total Return        left 95%	-1.49%	-1.63%	-9.25%	-2.24%	-5.10%	-0.45%	-19.74%	-11.29%	-28.29%
right 95%	35.41%	34.87%	45.31%	34.71%	34.87%	35.33%	70.68%	61.95%	49.75%
Excess Return     left 95%	-29.85%	-29.84%	-26.66%	-27.62%	-36.36%	-28.57%	-19.14%	-5.35%	0.00%
right 95%	42.31%	41.62%	41.25%	38.62%	44.67%	41.98%	48.61%	34.54%	0.00%
Notes:	tstat(row) = ave(row) / (std(row) / ((n - 1) ^ 0.5))
	68-95-97