The Chi-squared, t, and F distributions
[1] The Chi-squared (\chi^2) distribution
Definition
If X_{1},...,X_{p} are iid N(0,1) and X=(X_{1},...,X_{p})^T \sim N_{p}(0,I), (I is the pxp identity matrix)
V=\left \| X \right \|^2=X^TX (The squared length of X), then V = X_{1}^2+...+X_{p}^2 \sim \chi_{p}^2
[2] The t-distribution
Definition
Let X~ N(0,1) and V \sim \chi_{n}^2 be independent random variables,
then T= \frac{Z}{\sqrt{V/n}} \sim t_{n} with n degrees of freedom.
[3] The F distribution
Definition
Let U \sim \chi_{m}^2 and V \sim \chi_{n}^2 be independent random variables,
the variables W=\frac{U/m}{V/n} \sim F_{m,n} , the ratio of independent chi-squared variables.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment