Download Unit Root Process - Financial Econometrics - Exam and more Exams Econometrics and Mathematical Economics in PDF only on Docsity!
Semester 1 Examinations 2010/2011 (Autumn)
Exam Code(s) 1MIF Exam(s) M.Econ.Sc. (International Finance) Module Code(s) EC Module(s) Financial Econometrics I Paper No. 1 Repeat Paper External Examiner(s) Professor Cillian Ryan Internal Examiner(s) Professor John McHale Mr. Stephen O’Neill Instructions: Duration 3 hours No. of Pages 4 pages including this cover Department(s) Economics Course Co-ordinator(s) S. O’Neill Requirements : MCQ Release to Library: Yes √ No Handout Statistical Tables Graph Paper Log Graph Paper Other Material
Answer 4 questions only. Each question is worth 100 marks.
Please indicate clearly on the front of the answer book which questions
you have answered. If you have answered more than the required
numbered of questions please indicate which questions you would like
corrected – otherwise the first four answered questions in your answer
book will be corrected.
A1. Consider the following econometric model
Yt = β 1 + β 2 X2t + β 3 X3t + β 4 X4t + μt where Yt is the dependent variable, X2t, X3t and X4t are observations of independent variables, βi are unknown parameters and μt is a random error term. (a) Outline the principle underlying ordinary least squares regression in relation to the estimation to the above model. (25 marks) (b) Describe how you would calculate the goodness of fit of a regression equation. (25 marks) (c) The calculated R 2 for the above model is 0.94. Explain the meaning of this finding. (15 marks) (d) Explain how you would test a restriction on the above econometric model using an F test. (35 marks)
A2.
(a) Describe a unit root process. (25 marks) (b) The Dickey-Fuller test is a commonly applied procedure for testing economic time series for unit roots and stationarity. Briefly outline the Dickey-Fuller test procedure. (25 marks) (c) Explain what is meant by a random walk model, a random walk model with drift, a random walk model with trend and a random walk model with both drift and trend? (25 marks) (d) Often the disturbance terms in the Dickey Fuller test are not white noise. Suggest a remedy to this problem. (25 marks)
A5.
(a) What is autocorrelation? Which assumption of the classical linear regression model is violated by autocorrelation? (25 marks) (b) What are the possible causes of autocorrelation? (25 marks) (c) Describe how you would test for autocorrelation. (25 marks) (d) Explain how the problems associated with autocorrelation can be resolved? (25 marks)
A6.
(a) Outline what is meant by granger causality? (25 marks) (b) Describe how you would use a VAR model to test for Granger Causality between two stationary variables yt and xt. (40 marks) (c) VAR modelling has many advantages over “traditional structural models”. Describe these advantages and also some of the drawbacks of VAR modelling. (35 marks)
A7.
(a) Explain how an ARCH model can be used to model volatility in financial data series. (20 marks) (b) Explain how you would test whether an ARCH(p) model is necessary. (25 marks) (c) Why in recent empirical research have researchers preferred GARCH(1,1) models to pure ARCH(p) models. Explain your answer. (30 marks) (d) Outline the rationale for a GARCH in Mean model. Explain the model. (25 marks)
