Econometric Modelling
MODULE - 4
Dr. Sujata Kar
Assistant Professor
DEPARTMENT OF MANAGEMENT STUDIES IIT ROORKEE
1
Part 1: Introduction to Econometrics
Part 5: Univariate Time Series Modeling
Module 1: An Overview
Module 25, 26, 27: Problem of Serial Correlation
Module 2: Formulation of Econometric Modelling
Module 28: AR, MA & ARMA Processes
Module 3 & Module 4: Review of Basic Concepts
Module 29: Modelling Seasonal Variations
Module 5: Types of Data
Part 2: Overview of Classical Linear Regression Model Part 6: Models with Binary Dependent and Independent
Module 6 & 7: Simple Regression Variables
Module 8: Assumption of Classical Linear Regression Module 30 & 31: Spline Function & Categorical Variables
Module 9: Properties of OLS Estimators Module 32 & 33: Probit, Logit and Multinomial Logit Models
Module 10: Hypothesis Testing
Part 3: Multiple Regression Analysis & Diagnostic Tests
Module 11, 12 & 13: Multiple Regression Part 7: Multivariate Models
Module 14: Problems of Multicollinearity Module 33 & 34: Simultaneous Equations System
Module 15 & 16: Omitted Variables & Parameter Stability Module 35 & 36: Introduction to VARs
Module 17 & 18: Problem of Heteroscedasticity
Part 4: Statistical Inference and Hypothesis Testing
Module 19: t-test Part 8: Modelling Long Run Relationships
Module 20 & 21: Wald test Module 37, 38 & 39: Stationarity & Unit Root Testing
Module 22 & 23: F-test Module 40: Basics of Cointegration
Module 24: Chow test
Review of Basic Concepts
MODULE - 4
3
Sample Covariance
• Sample covariance is a measure of association between two
variables.
• In general, given n observations on two variables X and Y, the
sample covariance between X and Y is given by
• The sample covariance can be positive or negative depending on a
positive or negative association between X and Y.
4
Example: Consumption Function
The covariance
between PFCE
and GDP is
1131.54.
This positive
association can
be explained
using the
diagram.
5
Basic Covariance Rule
Rule 1 If Y = V + W, Cov (X, Y) = Cov (X, V)+ Cov (X, W)
Rule 2 If Y = bZ, where b is a constant and Z is a variable,
Cov (X, Y) = b Cov (X, Z)
Rule 3 If Y = b, where b is a constant, Cov (X, Y) = 0.
6
Population Covariance
• If X and Y are random variables, the expected value of the product
of their deviations from their means is defined to be the
population covariance,
• If X and Y are independent, their population covariance is 0,
because
and
7
Sample Variance
• For a sample of n observations, X1, ..., Xn, the sample variance
will be defined as the average squared deviation in the sample
• This is an unbiased estimator of population variance; proof
discussed in later modules.
• The variance of a variable X can be thought of as the covariance
of X with itself.
8
Variance Rules
Variance Rule 1 If Y = V+W, Var(Y) = Var(V)+ Var(W)+2Cov(V,W)
Variance Rule 2 If Y = bZ, where b is a constant, Var(Y) = b2Var(Z)
Variance Rule 3 If Y = b, where b is a constant, Var(Y) = 0
Variance Rule 4 If Y = V+b, where b is a constant, Var(Y) = Var(V)
9
Correlation Coefficient
Population correlation between X and Y,
• If X and Y are independent, ρXY will be zero since population
covariance, σXY will be zero.
• If there is a positive association between them, σ , and hence
XY
ρXY, will be positive with a maximum value of 1 with an exact
positive linear relationship.
• Similarly, if there is a negative relationship, ρXY will be negative,
with minimum value of –1.
10
Correlation Coefficient
• Sample correlation coefficient is obtained as
• Like ρ, r has maximum value 1, which is attained when there is a
perfect positive association between the sample values of X and Y.
• Similarly, it has minimum value –1, attained when there is a perfect
negative association.
• A value of 0 indicates that there is no association between the
observations on X and Y in the sample.
• Of course the fact that r = 0 does not necessarily imply that ρ = 0 or
vice versa.
11
Covariance versus Correlation
• Correlation coefficient is a better measure than covariance
because it does not depend on the scale of measurement.
• In our consumption function example, the variables PFCE and
GDP at market price, are measured in terms of billions of
Rupees.
• When the observations are divided by 1000, the covariance
value reduces from 1131543497 to 1131.543
• However, the correlation coefficient remains same at 0.99
12
References
• Dougherty, Christopher (2001). Introduction to Econometric.
Oxford, England.
• Brooks, Chris (2008). Introductory Econometrics for Finance.
Cambridge University Press, New York.
13
Thank You
14
