An algorithm to assess your investment performance

I’ve been quiet because I am in the process of writing up a critique which I hope to post shortly. The main argument is that most of the studies on individual investor behavior are potentially wrong, in two different way. Stay tuned.

However, the other side of the project is improving individual investor performance. Individual investors are often assessed using metrics they’ve never heard of  – like the Fama-French 3-Factor model.  But in my experience, even simpler metrics are unavailable to individual investors. For example, in my own online investment account my return information is available only on individual securities, and it is unadjusted for:

  • time the investment has been held
  • dividends received
  • volatility taken on
  • what the markets did during the same period

To get a feeling of alpha, I have had to reconstruct these figures myself using an unwieldy excel sheet, and they’ve helped me to understand my performance quite a bit more.

To that end, I’m building an algorithm in R, which will be web-hosted. You will be able to upload a set of transactions, and it will automatically calculate a variety of performance metrics and summary statistics. I hope to include tests of behavioral biases, such as the disposition effect.

Any thoughts or suggestions?


Leave a Reply

Your email address will not be published. Required fields are marked *