Data Science Classroom Series – 19/Nov/2021

Hypothesis Testing contd

  • Exercise: Some organization has designed a weight loss program.
    • You are give a sample of people who did this program.
    • Calculate the difference before and after
    • State the null hypothesis
    • Calculate the appropriate statistic
    • Decide if this is a one side or a two sidex test
    • Based on p-value decide at 1%, 5%, 10% significance if this program is working
    • The Organization claims the Weight loss program to be working
    • Refer Here for the dataset
    • The results are as shown the following Preview

Hypothesis testing for independent samples with known variance

  • The Dean of the University claims that since Engineering is difficult course the percentage mean of engineering students will be less than Management students by 4%
  • Refer Here for the dataset
  • Null Hypothesis H0=> μ(E) – μ(M) = -4%
  • Alternavite Hypothesis H1=> μ(E) – μ(M) != -4%
  • Z-Score Preview Preview
  • Analysis Preview

Hypothesis testing for independent samples with unknown variance but assumed to be equal

  • Refer Here for the dataset
  • We want to prove that prices in NY and LA are different
  • Null Hypothesis H0 => μ(NY) = μ(LA) => μ(NY)-μ(LA) = 0
  • Alternative Hypothesis H1 => μ(NY)-μ(LA) != 0
  • Calculate the pooled variance
  • T-Score Preview
  • Analysis: We reject the null hypothesis Preview
  • Rule: Reject the null hypothesis when T-Score is greater than 2

Note: For formulas Refer Here 12 Nov Classroom Notes

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

About learningthoughtsadmin