Back testing is a great tool to have when analyzing investment opportunities. But always remember the old jungle saying – “Historical performance is not indicative of future results”.
Therefore, investors should exercise caution when using back tested results.
The problem with back-test is that it is not a controlled experiment. This is because we are working with one specific realization of the time series and do not have the flexibility to change the environmental variables to derive a new time series to conduct an independent back-test.
Simply put, back tested results are not to be confused with what could happen in the future. At best, back testing will give you an idea of whether you are in the right direction or not. But it will not (and should not) lead you to your destination. Let it be one of many tools in your investment research kit.
At present, we only support historical back test as an evaluation tool (for future performance) for instruments, screens, signals, and baskets. And as we have seen above, historical back test is not the most robust way to evaluate future performance.
A lot of academic research has gone into building more robust evaluation frameworks for investment strategies, and it would be a shame if these are not introduced on our platform. Therefore, needless to say that we are actively working on launching more evaluation tools that could help investors make more informed decisions about their investment strategy.