A null hypothesis is a statistical hypothesis which suggests there is no statistically significant relationship among two variables under a study. Let us understand what is a null hypothesis and how does it work?

Example of Null hypothesis
Suppose a gambler wants to calculate the chance of the game being fair or not. If the game goes fair, then both the players gain expected earnings starting from 0. But if the game is not fair, then one player will get an expected earning to be positive and the other negative. The possibility of the fairness of the game is then calculated, where the gambler collects the earnings data from many repetitions of the game. Then, he calculates the average of these earnings from the data he collected. Later he tests the null hypothesis of whether the expected earnings were fair.
Suppose the average of the earnings was much different from zero, then he will avoid the null hypotheses and do the next alternative. But if the average amount is near zero, he will not reject the null hypothesis.

How a Null Hypothesis Works
The data is collected to prove whether the observations are true or false. Let us consider the example of plant growth. According to the Null hypothesis, the statement can be - the rate of growth of the plant in the presence of light is not affected by sunlight.
If the Null hypothesis is rejected, then further experiments are done to see that the relationship between two variables does exist. But the rejection of the null hypothesis doesn’t mean that the experiment was a failure and did not give the required results. But this helps to think about further experiments to be done.
The null hypothesis is denoted by H0, while the other hypothesis is written as HA or H1.

When is the Null Hypothesis rejected?
The p-value approach is used to accept or reject a null hypothesis. If the p-value is less than or equal to your significant level, then there should be a rejection of the null hypothesis. Alternatively, if the p-value is greater, the null hypothesis is not rejected.

Different types of Hypothesis
The different types of hypotheses are:
•    Simple Hypothesis: It specifies the population distribution wherein the method and the sampling distribution are the functions of the sample size.
•    Composite Hypothesis: In this hypothesis, the population distribution is not completely specified.
•    Exact Hypothesis: It specifies the exact value of the parameter. For example, μ= 50
•    Inexact Hypothesis: This kind of hypothesis does not specify the exact value of the parameter but gives a specific range or interval. For example, 45< μ <60.

#nullhypothesis #alternatehypothesis #statistics #mathematics #statisticsprobability #datascience #significance #nullhypothesisexperiments #hypothesisrejection #hypothesistesting

What is a Null Hypothesis?