GHANA INSTITUTE OF MANAGEMENT AND PUBLIC ADMINISTRATION (GIMPA) HOOL OF LIBERAL ARTS AND SOCIAL SCIENCES DEPARTMENT OF ECONOMICS, 2010/2021 END OF FIRST SESSION EXAMINATIONS, TITLE OF PAPER: ECO 708B: ECONOMETRICS FOR POLICY/ ECO 705B: ECONOMETRICS I LEVEL: MASTER LEVEL STUDENTS IN ECONOMICS DEPARTMENT DATE: SUNDAY, 2? APRIL, 2023 EXAMINER: DR WILLIAMS OHEMENG™ TWO AND HALE (2 42) HOURS TIME ALLOWED: INSTRUCTIONS Close book exam Answer ALL questions. Supply the examiner with only one answer to each question. Alll cell phones must be switched off for the duration of the test - note ~ not on silent, but Rene switched off. 5. You may not communicate by any means with any person ir or outside the exam venue for the duration of the exam. ~- INDEX NUMBER.[30 marks} QUESTION ONE: PART A: [10 marks] ti ta of ‘The following, summary statistics are generated from the production data of 22 wy = 100,8 = 10, (ay - 2)? = 60 Gi - DO. -Y) = 30 firms y= 20)" where y = In output and x = labour hours input (i) Compute the least squares estimates.of @ and B in the model, ysatpxte (ii) Estimate the output elasticity of labour (f ! (iii) Test the hypothesis that 6 = 1 (iv) Form a 95% confidence interval for 62, the variance of € . Note: 279,925,29 = 34.1696 and x? )575.09 = 9.5908 PART B: [10 marks] Consider the following regression model; ¥, = 24.7747 + 0.9415X,; + 0.0424X, SE = (6.7525) (0.8229) (0.0807) R? = 0.9635 R= 0.9531 df=7 RSS = 8,565.5541 ESS = 324.4459 where X, = income; X3 = wealth andy = consumption ae ~~-=—=(i) Calculate the t-values of each of the parameters of the model. (ii) How would you interpret the estimated parameters of the model? (ii) Is there multicollinearity in the regression? How do you know? Page 2 of 6PART C: [10 marks} Suppose the rel. tionship between consumption ¢ of a representat xpenditures and household income ive sample of households in G hana is specified as, Bx + B2X, + &, where i refers to household =1,2, ..., 1 i= It is believed that while low. consumption habits, consumption habits -income households have similar and less flexible high income houscholds have rather flexible and varied If one thinks that the consumption pattern so described corresponds to a situation where at least one of the fundamental as tumptions to the General Linear Regression Model is violated, then: (). Identify the fundamental hypothesis or hypotheses that could be systematically violated (ii). Indicate the consequences of the violation of those fundamental hypotheses on the ordinary least squares estimation of the specified model Suggest an alternative estimation approach which corrects for the identified violation. meme Page 3 of 6[20 marks] QUESTION TWO a et (2; tical 8s d through the following Lot {yz} and {2) denote two stochastical processes related thro t (4) and {z,} denote two stochas system: = 10 4+ Yvert N2%e-1 FF Ye = yo — Prate + VrYe-1 + 1220-1 © 9 - fe 24 = Dag — Bare + YasYe-1 + Yaa%e-1 + Fat where Eye and 1 are independent white noise Proce a structural VAR and a VAR in standard (i.e. into the standard form. (Use [4 Marks] (a) Explain the difference between a struc v reduced) form. Transform the above model i matrix notation where convenient). (b) Explain the identification problem and the meaning of the coefficients b,and b2,. How can we solve the identification problem? [4 Marks] (©) Under which conditions ate y, and/or Zp stationary? [4 Marks] ( Explain Granger causality and exogeneity and the difference between these two concepts, How could you test for exogencity? [4 Marks] : variables might be cointegrated and want to ‘est for it. How many cointegrating vectors can there be at most? [4 Marks} OTe Ou tbe ee (©) Your conjecture that the above the vec Page 4 orgQUESTION THREE Consider the following system of two equations [20 marks] Vie = Go + Boe + AX ye + WyX yp $ Uy eecceeeeee (1) Yar = Bo + BrYie + BoXy_ + Uy eee 4 (2) (a) Explain, with reference to these equations, the undesirable consequences that would arise if (1) and (2) were estimated separately using OLS. [4 Marks] (b) What would be the effect upon your answer to (a) if the variable y,, had not appeared in (2)? [3 Marks} (c) State the order condition for determining whether an equation which is part of a system is identified. Use this condition 10 determine whether (1) or (2) or both or neither are identified. [3 Marks] (d) Explain whether indircet least squares (ILS) or two-stage least squares (2SLS) could be used to obtain the parameters of (1) and (2). Describe how each of these two procedures (ILS and 2SLS) are used to calculate the parameters of an equation. Compare and evaluate the usefulness of ILS, 2SLS and IV. [6 Marks] (e) Explain briefly the Hausman procedure for testing for exogeneity. [4 Marks] “4QUESTION FOUR [20 marks] a) Outline step-by-step, the Johansen’s methodology for testing, for cointegration between a set of variables in the context of a VAR. b) Suppose that a researcher has a sgt of three variables, ye(t= Aya ouensT), ie. Jy denotes a p-variate, or p X 1 vector, that she wishes to test for the existence of cointegrating relationships using the Johansen procedure. What is the implication of finding that th a value of; } - ¢ rank of the appropriate matrix takes on i) 0 ii) 1 ii.)2 iv.)3? ©) The researcher obtains results for the Johansen test using the variables outlined in part (b) as follows: r Amax 5% critical value 0 38.65 30.26 i 26.91 23.84 2 10.67 17.72 a BSB ae a / ec Determine the number of cointegrating vectors, explaining your answer.