 # Hypothesis Testing

Apologies for rookie question ,

• Can someone please explain the intuition behind the rejection of null hypothesis when statistic lies in rejection region?
• Why do we reject null hypothesis in case of low p value?
• P value is basically a conditional probability, which assumes that null hypothesis is true. (Please refer an example shown in the lectures for better understanding of this point).
• We tend to reject the null hypothesis, in case this probability is fairly less, as it clearly signifies that our conditional statement (null hypothesis) is false.
1 Like

P-value: What is the probability of getting a test statistic of that magnitude or high when we assume that the null hypothesis is true? i.e. if nothing was going on, how unusual to see this estimate?

E.g. We would have run a sales campaign and found that there was an increase in sales by 5%. Was this increase due to sales campaign or by chance? To know this, we will have to test this hypothesis by dividing into test and control groups. Null hypothesis will assume that there was no impact on sales whereas alternate hypothesis would assume that there was an impact in sales between these two groups. We will then calculate the p-value. If the p-value was 0.20, then there is a 20% probability that the null hypothesis was true even though there was an increase in sales. Hence, we would always like have p-value low.

1 Like