Haugen Model Performance

U.S. Model's Strategies

How We Demonstrate Performance

In order to test and demonstrate the performance of the quantitative data that we provide to our clients, Haugen Equity Signals constructs and back-tests 50-stock portfolios, and then rebalances these portfolios each quarter.

Here’s how these 50-stock portfolios are constructed:

  •  We begin with a population of the top 2,000 U.S. stocks by market cap. After calculating expected returns on these, we rank the stocks from lowest to highest expected return.
  •  Next we filter out stocks that are priced at under $5 per share or have a market cap under $300M. In the tests shown here, we imposed no constraints for sector weighting.
  •  From the remaining population, we select the best 50 stocks for each strategy (the highest expected return stocks for the long-only strategies; and the highest and lowest expected returns for a market neutral strategy).

At the end of the month we rebalance each portfolio. Based on newly-generated expected returns, we replace up to 50 stocks in each portfolio (i.e. 100%).

All back-tests use date-stamped data to eliminate the possibility of look-ahead bias. We include a benchmark in each chart for comparison purposes.

To see portfolio returns for each strategy, choose from the menu to the left.

For more details, click on the terms at the bottom of each chart.


ENHANCED MODEL'S LONG-ONLY STRATEGY
Year Long Backtest
Total Return
S&P 500 Index
Total Return
Long Backtest
Excess Return
1996 35.71% 22.96% 12.75%
1997 66.59% 31.77% 34.82%
1998 41.23% 24.13% 17.10%
1999 69.14% 20.89% 48.25%
2000 12.73% -7.46% 20.19%
2001 -1.48% -11.46% 9.97%
2002 -4.37% -21.55% 17.18%
2003 52.00% 31.04% 20.95%
2004 16.15% 11.95% 4.21%
2005 36.22% 6.13% 30.09%
2006 19.12% 15.74% 3.38%
2007 34.05% 5.13% 28.92%
2008 -20.43% -37.31% 16.88%
2009 73.65% 28.33% 45.32%
2010 29.02% 16.93% 12.09%
2011 8.32% 1.02% 7.30%
2012 13.74% 16.43% -2.69%
2013 46.40% 33.57% 12.83%
2014 14.37% 12.56% 1.82%
2015 -8.67% 0.47% -9.14%
2016 21.10% 12.76% 8.34%
2017 19.73% 21.13% -1.40%
2018 -1.41% -5.24% 3.84%
2019 35.91% 31.02% 4.89%
2020 20.02% 5.58% 20.02%
Average Linked Annual Ret. 22.74% 9.22% 13.70%
Average Annual Return 26.11% 10.00% 16.22%
Longitudinal Std.Dev. 25.33% 17.84% 14.75%
Sharpe Ratio (20-yr) 0.89 0.53 0.48
T-Stat for the Mean Return 4.72 2.57 4.79
Probability Total (Excess) Ret < 0 15.16% 28.80% 13.59%

Cumulative Returns

1 year

Image

3 year

Image

5 year

Image

10 year

Image

Decile Performance

In the charts on the pages below showing returns for the Europe, U.S. and Japan, decile returns are plotted against decile rankings and a slope is computed for the line of best fit through the 10 plot points.
You will find the annual, quarterly, and monthly returns for each of the deciles as well as the returns to the benchmark index, the slope of the line of best fit, and the r-squareds showing the precision of fit around the line. These results are based upon a back-test of our most recent models. We use a date-stamped database to avoid "look-ahead" bias. In addition, firms that do not currently exist are used in the modeling process to avoid survival bias.

Verified Results

Investars.com has audited our monthly results for more than ten years against about 20 other research organizations. Haugen Equity Signals performance for the last six years (the longest period Investars shows) is ranked #1.

At Investars, performance results are segmented by type of firm, the universe of stocks covered, sectors, as well as long and short picks. The top five firms in each category are displayed.

Directions for use:

  1. From their home page, select “ranks“ from the top menu
  2. Investars audits data from different types of financial firms. To see our company, change the drop-down box that shows “Fundamental” to “quantitative”.
  3. Then to show the longest historical performance, change the date range from ”1 year (trailing)” to ”6 years trailing"

Predictive History

Expected Return History Files

The history files include expected returns from our current models. The populations include:

  • US: between 3,000 and 4,500 stocks
  • Europe: between 3,000 and 3,500 stocks
  • Japan: between 450 and 3,500 stocks
  • Emerging Markets: between 800 and 3,200 stocks

The data goes back to 2003 for each model. All data are date stamped and represents what was actually sent out to our clients at the time. If the model was not in production in 2003, the data before the inception date represents backtested data, while all data on or after the inception date is the data we actually sent to our clients.

Dates of inception

  • U.S. Standard: before 2003
  • U.S. Enhanced: April, 2006
  • Europe: before 2003
  • Japan: before 2003
  • Emerging Markets: July, 2015

The data are intended to be used by prospective clients in order to test their own screening methods as well as their own trading strategies. Our returns do not take into consideration risk, but some of our clients have successfully utilized risk management tools when building their portfolios, which may also be employed in back-testing this data.