Econometrics - Functional Form - Chapter 5 Gujarati - 2020

This video is based on Chapter 5 of D.N. Gujarati & Porter’s : Essentials of Econometrics. The Topic discussed is the various types of Functional Forms in which the variables can be expressed in regression analysis, which better fits the data as well as supports the theoretical understanding of the subject.
I have discussed the following issues:
1. Types of Functional forms (Non-Linear in Variables) can be there (not an exhaustive list) – Yet in domain where we have linearity of Parameters

2. Log-Linear Model or Log - Log Model or Double Log Model or Constant Elasticity Model: Both Dependent as well as Independent Variables are in Log form and B2 Coefficient represents the elasticity.

3. Semi – Log Models :
a. Log-Lin Model or Growth Rate Model : where dependent variable is in log form and the independent variable is in linear form & B2 represents percentage change in Y divided by change in X. We need to multiply the Beta by 100 to get correct interpretation.

b. Lin-Log Model : where dependent variable is in linear form and the independent variable is in log form & B2 represents change in Y divided by percentage change in X. We need to divide the Beta by 100 to get correct interpretation.

4. Reciprocal Model: Where the explanatory variable : X comes in reciprocal form – the independent variable asymptotically approaches B1 & based on positive or negative values of Beta Coefficients we can have 4 types of cases (3 economically viable) : Case of AFC, Case of Engel Curve, Case of Phillips Curve.

5. Polynomial Models: Where the same explanatory variable enters multiple times with different powers. The case of Total Product, Average cost curves are examples of such use cases.

6. Regression through Origin: Model in which there is no intercept: the calculation of parameters changes from deviation form to absolute form, the R2 can be negative & sum of errors need not be zero (Prove all 3 cases)

7. Regression of standardised variables: this done to suppress the wide variation and different scales on which the variables are calculated. This model as well is a case of regression through origin (Prove how)

This Chapter is extremely important from the point of examination for Semester IV students of Economics Hons. & BBE of Delhi University. While students of Masters, MBA, UGC-NET, IES Exam would also find it useful in their course of preparation.

My Best Wishes & Happy Learning😊
The Pink Professor
Siddharth Rathore
Assistant Professor
Department of Economics
Gargi College
University of Delhi