What is Type I and Type II Errors?

Let’s first understand NULL(H0) and ALTERNATE(HA) hypothesis:

• Here the assumption needs to be made for population, not sample.
• Initially we assume that the null hypothesis is true.
• Based on outcome of statistical tests, either we “reject the null hypothesis” or we “failed to reject the null hypothesis”.

Type I and type II Errors

• Type I Error: Rejecting the null hypothesis when it’s true
• Type II Error: Not rejecting the null hypothesis when it’s false

• Here our goal is to control both the errors simulteneously by maximizing the power of a test by holding (α) to be low.
• For example we controll type I error at 5% that means when α=5%, our analysis has been done at the confidence level of 95%.