# 4.3.4 Bias

###### 4.3.4  Bias

The bias of an estimator H is the expected value of the estimator less the value θ being estimated:

[4.6]

If an estimator has a zero bias, we say it is unbiased. Otherwise, it is biased. Let’s calculate the bias of the sample mean estimator [4.4]:

[4.7]

[4.8]

[4.9]

[4.10]

[4.11]

where μ is the mean E(X) being estimated. The sample mean estimator is unbiased.

###### 4.3.5 Standard error

The standard error of an estimator is its standard deviation:

[4.12]

Let’s calculate the standard error of the sample mean estimator [4.4]:

[4.13]

[4.14]

[4.15]

[4.16]

[4.17]

[4.18]

where σ is the standard deviation std(X) being estimated. We don’t know the standard deviation σ of X, but we can approximate the standard error based upon some estimated value s for σ. Irrespective of the value of σ, the standard error decreases with the square root of the sample size m. Quadrupling the sample size halves the standard error.

###### 4.3.6 Mean Squared Error

We seek estimators that are unbiased and have minimal standard error. Sometimes these goals are incompatible. Consider Exhibit 4.2, which indicates PDFs for two estimators of a parameter θ. One is unbiased. The other is biased but has a lower standard error. Which estimator should we use?

Exhibit 4.2: PDFs are indicated for two estimators of a parameter θ. One is unbiased. The other is biased but has lower standard error.

Mean squared error (MSE) combines the notions of bias and standard error. It is defined as

[4.19]

Since we have already determined the bias and standard error of estimator [4.4], calculating its mean squared error is easy:

[4.20]

[4.21]

[4.22]

Faced with alternative estimators for a given parameter, it is generally reasonable to use the one with the smallest MSE.