t - Distribution (Student t - Distribution)

From MM*Stat International

Jump to: navigation, search

t-Distribution is also known as the Student t-Distribution. If Z has a standard normal distribution N(0;1) and Y, the sum of squared standard normal random variables, has a -distribution with degrees of freedom , then we define as the t-distribution with parameter (shortly written as t()), if Z and Y are independent. The parameter represents the degrees of freedom for the random variable Y. The random variable T has range and expected value and variance: The following diagram plots the density function a t-distribution for different numbers of degrees of freedom .

Nl s2 29b 2.gif

. The Chi-square, t-, and F- distributions are distributions that are functions of Normal random variables that are particularly useful in statistics. On the t-Distribution. The density function of a t-distribution is a bell-shaped symmetricdistribution with expected value (as a standard Normal distribution). However, a t-distribution has heavier tails than a standard Normal distribution. In other words, the t-distribution will be more dispersed than a standard Normal distribution. The variance of the standard normal distribution is 1, but the variance of a t-distribution equals (for ). As , the density function of the t-distribution converges to the standard normal distribution. For , a Normal distribution can produce a good approximation to a t-distribution. The t-distribution is tabulated different values of .