Robeco uses cookies to analyze your visit to this site, to share information via social media and to personalize the site and advertisements in line with your own preferences. By clicking on agree or by continuing on this site, you agree to the above. More information and adjusting cookie settings.

Robeco uses cookies to analyze your visit to this site, to share information via social media and to personalize the site and advertisements in line with your own preferences. By clicking on agree or by continuing on this site, you agree to the above. More information and adjusting cookie settings.

# How do you assess the track record of a strategy?

25-02-2015 | Insight How do you assess the track record of a strategy in order to ensure the best possible future returns? The solution to this problem was discovered by former Guinness employee, William Sealy Gosset. He needed a cheap way to test the quality of barley for brewing beer and devised the t-value to tackle the problem of drawing conclusions based on a small sample size using a formula known as the t-statistic.

The formula can be applied to many industries, including finance, because investors often face a similar problem of not having an unlimited sample size. The t-value can be used to test the statistical significance of excess returns - the returns of a strategy relative to an index. The standard deviation of these returns is the tracking error (TE). Once the t-value is calculated, it can be compared to the t-statistic for T-1 degrees of freedom and a 95% significance level. The outcome of this can be determined by the standard t-tables.
If you apply this formula to Robeco Core Developed Markets* (Institutional Global Enhanced Index fund), which has an average annual outperformance since inception relative to the MSCI World of 1.17%, a TE of 1.09%, and IR of 1.08 and a >10-year track record, the t-value is 1.09*√10.1= 3.46. This is statistically significant; because the critical t-statistic for 95% significance is 1.86.

* The value of your investments may fluctuate. Past results are no guarantee of future performance.