Question

You have been analyzing stock prices for a while, and you have come to believe that you may have discovered a new asset pricing anomaly. While a behavioural explanation for it is plausible, what caveats should you keep in mind?

Answer 1

Attempting to apply anomalies is a risky way of reaching to conclusion. Many anomalies are just illusionary in the first place, but they are also unpredictable. They are often result from intense large-scale data analysis that looks at portfolios consisting of hundreds of stocks that deliver just a fractional performance advantage.

• Keep in mind, it has to be consistent over a long period of time and over a large mixture of variables.
• On the abstract level detection of the anomalies seems like a simple task. But this task can be very challenging.
• Normal observations could be considered as anomalies and vice-versa.
• A systematic miscalculation may seem like an anomaly.
• Irrational approach over the detection of anomaly could be possible.
• Training and validation data availability for proving the model is a huge problem.