Two types of errors are defined for the p-value; these errors are given below:

• 1. Type I error
• 2. Type II error

Type I Error:

It is defined as the incorrect or false rejection of the Null hypothesis. For this error, the maximum probability is alpha, and it is set in advance. The error is not affected by the sample size of the dataset. The type I error increases as we increase the number of tests or endpoints.

Type II error

Type II error is defined as the wrong acceptance of the Null hypothesis. The probability of type II error is beta, and the beta depends upon the sample size and value of alpha. The beta cannot be determined as the function of the true population effect. The value of beta is inversely proportional to the sample size, and it means beta decreases as the sample size increases.

The value of beta also decreases when we increase the number of tests or endpoints.

We can understand the relationship between hypothesis testing and decision on the basis of the below table:

Importance of P-value

The importance of p-value can be understood in two aspects:

• • Statistics Aspect: In statistics, the concept of the p-value is important for hypothesis testing and statistical methods such as Regression.
• • Data Science Aspect: In data science also, it is one of the important aspect Here the smaller p-value shows that there is an association between the predictor and response. It is advised while working with the machine learning problem in data science, the p-value should be taken carefully.