# Removing stocks from simulation based on long term out of sample performance

I have performed a simulation on a stock universe and have found some stocks that out of sample have never performed (every day they always lose money in the simulation). I don't want to introduce any look ahead bias but these stocks have consistently lost money for 2 years (I simulate once a day). Can I remove them from the simulation?

Thanks

• It surprises me that $500 = 2 \times 250 \textrm{ trading days}$ returns are all negative. How are you simulating? – Bob Jansen Dec 4 '18 at 6:53
• Thanks Bob, The returns are not all negative but the trend is very definitely consistently negative. I'm testing in-sample (1 year) and rolling forward (by a few days) out of sample – Stacey Dec 4 '18 at 8:08
• In your question you state they "always lose money". Some stocks might perform bad, if you remove badly performing stocks from your sample you're definitly introducing look-ahead bias. – Bob Jansen Dec 4 '18 at 8:47

## 1 Answer

Dont remove them, this is a common topic. For example, when you download a set of data, most of the time, you only get companies that did not defaulted/delisted which create a strong bias. Common sense would say, ok if the stock is underperforming for 2 years, it might get delisted and the model should capture this information. Which again, happens in the real world, specially after many months with negative results.

• Thanks that is a sensible answer. Would the same argument hold instead of removing stocks you remove the sector instead? – Stacey Dec 4 '18 at 11:22
• Yes, e.g. look XLE. In the last 10 year it has a 10-Yr change % of -13.51 and has not been removed in any way. A) Models should capture underperforming sectors to 1. properly allocate assets and 2. forecast specific defaults. and B) An underperforming sector can be beneficial in a portfolio for reducing risk, variance, etc. depending on the correlations. – TomDecimus Dec 4 '18 at 12:48