3.10.3 Chi-squared Distributions
Suppose Z is a standard normal random variable. How is Z 2 distributed? The answer is a chi-squared distribution. More generally, let Z1, Z2, … Zν be ν independent standard normal random variables, and let δ1, δ2, … δν be constants. Then the random variable
[3.112]
has a chi-squared distribution with ν degrees of freedom and noncentrality parameter7
[3.113]
We denote a chi-squared distribution χ2(ν,δ2). If δ2 = 0, the distribution is said to be centrally chi-squared. Otherwise, it is said to be noncentrally chi-squared. The PDF for a central chi-squared distribution is
[3.114]
For noncentral chi-squared distributions, this generalizes to
[3.115]
where Γ( ) denotes the gamma function8
[3.116]
The PDF of a chi-squared distribution is illustrated in Exhibit 3.19.
The expectation, standard deviation, skewness, and kurtosis of a chi-squared distribution are
[3.117]
[3.118]
[3.119]
[3.120]
