Skip to main content
. Author manuscript; available in PMC: 2009 Apr 21.
Published in final edited form as: Neural Comput. 2008 Dec;20(12):2895–2936. doi: 10.1162/neco.2008.10-07-633

Figure 1.

Figure 1

Behavior of bias and variance of naive estimates of indices of concentration fα. (A) Two symbols, with probabilities p1 and p2 = 1 − p1. (B) Three symbols, with probabilities p1, p2, and p3 = 1 − p1p2. Column 1: fα, linearly rescaled into the range [0, 1]. Column 2: the coefficient of 1/N in the asymptotic bias (same normalization as column 1). Column 3: the coefficient of 1/N in the asymptotic variance (same normalization as column 1).