Chi-Square Distribution - Overview
Chi-Square Distribution - Overview
1. Are lottery numbers evenly distributed or are some numbers likely to occur with a greater
frequency?
2. Are the preferred movie genres different across different age groups?
3. Do coffee machines dispense approximately the same amount of coffee each time?
One can answer these questions by conducting commonly used statistical tests like goodness-of-fit, test
for independence, etc which use the chi-square distribution at their core.
We will be discussing the Chi-square Test of Independence in the upcoming video. Before we dive into
the hypothesis test, let us spend some time understanding the probability distribution that the test uses,
i.e., the chi-square distribution.
where is a random variable having a chi-square distribution, 's are independent standard normal
variables, and is the degrees of freedom of the chi-square distribution.
Just like the mean and variance are the parameters of a normal distribution (denoted
by ), the same way, the degrees of freedom is the parameter of a chi-square distribution
(denoted by ).
For a chi-square distribution , the mean and variance of the distribution are and respectively.
As the degrees of freedom increase, the chi-square distribution tends towards a normal
distribution (the density curve becomes more bell-shaped)
The plot below shows the change in a chi-square distribution with increasing degrees of freedom (also
commonly denoted as df).
We can observe from the above plot that as the value of degrees of freedom increases the curve reaches
close to a normal distribution.