Look-ahead Bias

Look-ahead bias is use of information outside the time frame of the simulation period. For example, survivorship bias can be one type of look-ahead bias in a decade long study that exclude bankrupt companies. Look-ahead bias corrections can neutralize results that extrapolate back tested historical portfolio results. If, for example, current quarterly earnings were not known when a simulation was run, look-ahead bias can be expected and better results found once the true earnings are known. Look-ahead bias is often cited in discussing hedge fund performance. Elimination or recognition of look-ahead bias for multi-period winner or portfolio performance persistence studies is an important factor in the accuracy of these studies. Look-ahead bias can also be an important factor in calculating or correcting fund liquidation forecasts.