Econometrics Notes Economics honours ppt

Econometrics

1. Binod Goswami’s Classes on Econometrics B.A. Fourth Semester Honours Topic- Chi-Square Test B asic Pervious knowledge required on- Hypothesis- Null and Alternative Errors and its types Parametric and Nonparametric Test Nominal and Categorical Variable 5/24/2021 1 BG's Classes on econometrics

2. Hypothesis- Null and Alternative Hypothesis is a statement about a population.Ex - Mean weight of students of NGC is 46 kg and its standard deviation is 9 It is an assumption made about parameter values. Validity of hypothesis is tested with the help of statistical techniques (CHI-square test, T- test,F -test) These test tells us whether hypothesis is to be accepted or to be rejected 5/24/2021 2 BG's Classes on econometrics

3. Types of Hypothesis – Null hypothesis and Alternative hypothesis Null hypothesis Null hypothesis-a statistical hypothesis stated to test its validity is called n ull hypothesis. Denoted by H H is always tested on the basis of sample results and is ultimately either rejected or accepted. It is the hypothesis of no difference. If the population mean is  then H is written as H :  =  ,  is a specified value of  Similarly if and  1 are  2 the parameters of two populations , the null hypothesis is written as H :  1 =  2 5/24/2021 3 BG's Classes on econometrics

4. Types of Hypothesis – Null hypothesis and Alternative hypothesis Alternative hypothesis- hypothesis accepted as a result of rejection of null hypothesis. Denoted as H 1 It is complementary to the null hypothesis if the null hypothesis is that the population has specified mean  (say) i.e. H :  =  Then the alternative hypothesis may be H 1 :    H 1 :  >  H 1 :  <  5/24/2021 4 BG's Classes on econometrics

5. H 1 :    Two tailed hypothesis H 1 :  >  One tailed – Right tailed H 1 :  <  One tailed – Left tailed 5/24/2021 5 BG's Classes on econometrics

6. Errors in Hypothesis Testing In hypothesis testing, the decision to accept or reject of null hypothesis is taken on the basis of the sample result. The decision taken in this way may not be correct always. Thus Errors arise. There are 4 possibilities when a statistical hypothesis is tested 5/24/2021 6 BG's Classes on econometrics

7. Reject H when it is actually true Type I error(  ) Accept H when it is actually false Type II error (  ) Accept H when it is actually true No error Reject H when it is actually false No error 5/24/2021 7 BG's Classes on econometrics

8. Type I and Type II errors Accept H Reject H H is true No error Type I error H is false Type II error No error 5/24/2021 8 BG's Classes on econometrics