Learning outcomes for participants:

At the end or during the course participants will be able:

(a) To conceptualise the vital concepts and issues of econometric analysis and modelling;

(b) To apply the econometric concepts and tools to understand and analyse their countries' economic behaviour;

(c) To understand the complex nature and inter-linkages among sectors and markets of their economies;

(d) To acquire skills and knowledge of econometric modelling for strategic thinking and understanding;

(e) To acquire methodological foundation necessary for future studies; and

(f) To synthesis of ideas, views and evidence.

    Career relevance:

The knowledge and skills obtained from the course will be useful for business mangers and development practitioners in analysing and maximising the effectiveness of polices and business decision. Aware of econometric tools are necessary for effective decision making in business and policy administration/management.

    Pedagogical techniques:

(a) Class room lectures;

(b) Computer workshops and in class discussion;

(c) Reading assignments; and

(d) Homework assignments and group discussion.

Econometrics has become one of the most important tools for policy makers of the day in quantifying the impacts of various policies on their economies. To tackle practical problems, policy makers and business managers need to build econometric models. The objectives of the course are: (a) to understand important econometric tools/techniques; (b) to learn simple models; and (c) to use econometric techniques in real world policy analysis.

This course is also designed to provide the practical illustration of techniques used in applied macroeconometric. This course is highly recommended for students who intend to do quantitative analysis (using time series data in particular) in their thesis writing.

Assessment will depend on the performance of the followings:

(a) Class participation/commitment/contribution etc 10%

(b) Homework assignments (3-5) 30%

(c) Examination 60%

1. Topics to be discussed in the course

(1) Introductory lecture:

a. What is Econometrics?

b. Why separate discipline?

c. Methodology of Econometrics

d. Types of Econometrics

e. Mathematical and statistical prerequisites

f. Illustrative examples

(2) The nature and sources of data for economic analysis

a. Types of economic data

b. Obtaining data

c. Working with data

(3) Correlation

a. Understanding correlation

b. Why variables are correlated

c. Correlation between several variables

(4) Single-equation regression model

a. The nature of regression analysis

b. Regression verses causation

c. Regression verses correlation

d. Illustrative examples

(5) Two variable regression analysis

a. The method of ordinary least squares

b. The classical liner regression model

c. Standard errors of least-squares estimates

d. The coefficient of determination and other statistics

e. Illustrative examples

(6) Hypothesis testing

a. General comments

b. The confidence-interval approach

c. The test-of-significance approach

d. Some practical aspects

e. Illustrative examples

(7) Extensions of two-variable linear regression model

a. Regression on standardized variables

b. Functional forms of regression models

c. How to measure elasticity: Log linear model

d. Semi log models

e. Illustrative examples

(8) Dummy variable regression models

a. The nature of dummy variables

b. Some technical aspects of the dummy variable model

(9) Regression with time lags: Distributed lag models

a. Why we need lags?

b. Selection of lag order

c. Estimation of equation with lags

Topics from 10 to 12 are optional. We will discuss these topics if time allows to do so.

(10) Multicollinearity

a. What happens if the regressors are correlated?

b. The nature of multicollinearity

c. Theoretical consequences of multicollinearity

d. Piratical consequences of multicollinearity

e. Detection of multicollinearity and remedial measures

(11) Heteroscedasticity

a. What happens if the error variance is non-constant?

b. The nature of heteroscedasticity

c. Consequences of using OLS in the presence of heteroscedasticity

d. Detection of heteroscedasticity

e. Remedial measures

(12) Autocorrelation

a. What happens if the error terms are correlated?

b. The nature of the problem

c. OLS estimation in the presence of autocorrelation

d. Consequences of using OLS in the presences of autocorrelation

e. Detecting autocorrelation

f. Remedial measures