Basic Econometrics — DU B.A (Prog.) Economics as Major Previous Year Question Papers

1. Consider the following regression output : where Y is labour force participation rate (LFPR) in 1972 and X is LFPR of women in 1968. The regression results were obtained from a sample of 19 cities of the country. (a) Some part of the printed output was damaged due to a printer error, leaving some values unreadable. Using the information that is still clear, determine the missing results. (b) Interpret the intercept coefficient, slope coefficient and R2 (c) The researcher expects the coefficient on X to exceed 1. Test the appropriate hypothesis to evaluate this expectation at 5 percent level. (6×3=18) 2. (a) State the assumptions of classical linear regression model. (b) A researcher estimates the model of the form where Y is the annual expenditure on food, D1 is 1 for female and 0 for male, and D2 is 1 for male and 0 for female. Explain the problem that she might encounter in estimation. (c) What are the consequences of multicollinearity? (6×3=18) 3. (a) Consider the following model : Yi = β0 + β1D1i + β2D2i + β3Xi + εi where Y is hourly wage, X is education in years, D1 is 1 for female and 0 for male, and D2 is 1 if non-white and 0 if white. Write down the mean wage function for each of the four groups of individuals classified by gender and race. (b) What is an interaction dummy variable? Explain why it is included in a regression equation, giving a suitable example. (c) Explain the problem of heteroscedasticity in the regression model. 4. (a) Why is stochastic error term included in the regression model? (b) Illustrate diagrammatically that Ordinary Least Squares estimators are unbiased and efficient. (c) State the steps involved in testing for positive serial correlation using Durbin Watson test. (6×3=18) 5. The Engel expenditure curve relates a consumer’s expenditure on a commodity to her total income. Let Y = consumption expenditure on a commodity and X = consumer income, consider the following models : Model III: Yi = β0 + β1 ln Xi + εi (a) Interpret the slope coefficients of Model I, II and III. (b) Illustrate the relationship between Y and X by drawing curves for the cases when β1 > 1, 0 < β1 <1 and β1 < 0 for Model II and for the cases when β1 < 0 and β1 > 0 for Model III. (c) What is the purpose of including an intercept term in the model? 6. (a) If the error term in the regression model is heteroscedastic, what are the implications for the estimation of coefficients? (b) Which assumption of the classical linear regression is violated when tolerance (TOL), reciprocal of Variance Inflation Factor (VIF), is equal to zero? Explain the nature of the problem. (c) What are the steps involved in Jarque-Bera test to examine the normality of errors? (6x3=18) ESS = 0.0358 RSS = 0.0544 (7) ARa 7750 fae fe: X, = Bo + B, D,i + B,Dzi + B,X, +E, VIŠT Y TIA e e e e {6 cà {áe{ qee {ey { d (Ordinary Least Squares) (bbed (efficient) aa ği STATISTICAL TABLES 519 Table B-1 Critical Values of the t-Distribution Level of Significance