TOPICS
Search

Biased Estimator


A biased estimator is an estimator which exhibits estimator bias.

Bias can arise when the observations included in an estimate are selected using the observed sequence. For example, consider three independent fair coin tosses, with heads and tails denoted H and T. Select each of the second and third tosses whose immediate predecessor is H. Let K be the number selected and S the number of selected heads. When K>0, use the empirical heads proportion p^^=S/K.

Of the eight equally likely sequences, TTT and TTH have K=0 and are excluded. On the other six, p^^ takes the value 0 three times, 1 twice, and 1/2 once. Its conditional expectation is therefore

 E[p^^|K>0]=(0+0+0+1/2+1+1)/6=5/(12).

Thus its conditional estimator bias for estimating the underlying heads probability 1/2 is 5/12-1/2=-1/12 (Miller and Sanjurjo 2018, Numberphile 2018).

Pooling the selected outcomes over the complete table of eight sequences instead gives four heads among eight selected tosses, a proportion of 1/2. This weights the sequences by their numbers of selected tosses, rather than giving each eligible sequence equal weight, and is a different statistic.


See also

Estimator, Estimator Bias, Unbiased Estimator

Explore with Wolfram|Alpha

References

Miller, J. B. and Sanjurjo, A. "Surprised by the Hot Hand Fallacy? A Truth in the Law of Small Numbers." Econometrica 86, 2019-2047, 2018. https://doi.org/10.3982/ECTA14943.Numberphile. "Is the 'Hot Hand' Real?" Apr. 20, 2018. https://www.youtube.com/watch?v=bPZFQ6i759g.

Referenced on Wolfram|Alpha

Biased Estimator

Cite this as:

Weisstein, Eric W. "Biased Estimator." From MathWorld--A Wolfram Resource. https://mathworld.wolfram.com/BiasedEstimator.html

Subject classifications