[SOLVED] statistic software theory Warwick Business School University of Warwick

Warwick Business School University of Warwick
MSc in Finance
MSc in Finance & Economics
IB93F0: Research Methodology and Dissertation Term 1, 2018/2019
Exercise # 1: Tests of Random Walk Hypothesis
Objective:
The objective of the project is to help students to learn how to gather, clean, analyse and present data; formulate a hypothesis; interpret results; and gain other transferable skills for their dissertations. An important aspect of this exercise is the economic interpretation of results and their discussion in the context of the strengths and weaknesses of the tests themselves. A key part of research is an ability to think independently and to distinguish what are the most important issues. Think about this and clearly identify what you wish to do before you start. If possible, make your work distinct by giving reasons in your introduction. The quality of the interpretation is as important as the empirical analysis and the results. Please keep in mind that this task is all open-ended: there is no guarantee that you end up having similar empirical results compared to what have been documented previously. As long as you are certain about your analysis, there is no point wasting time replicating the existing results. Just proceed with your findings and try to interpret them by relating them to the concepts covered in class. You can extend major results contained in the key readings. But please stick to the 1500-word limit. The project is due at 12noon on Thursday 30 May 2019.
Reference:
Campbell, J.Y., A.W. Lo and A.C. MacKinlay, 1997, The Econometrics of Financial Markets, Princeton University Press
Lo, A. and A.C. MacKinlay, 1988, Stock Market Prices Do Not Follow Random Walks: Evidence from a Simple Specification Test, Review of financial Studies 1, 41-66.
Lo, A. and A.C. MacKinlay, 1989, The Size and Power of the Variance Ratio Test in Finite Sample, Journal of Econometrics 40, 203-238.
Fama, E., 1970, Efficient Capital Markets: A Review of Theory and Empirical Work, Journal of Finance 25, 383-417
Fama, E., 1991, Efficient Capital Markets:II, Journal of Finance 46, 1575-1618 Data:
You need to choose ONE or TWO indices (e.g., VW and EW) to conduct tests for weak form efficiency using monthly returns; for simplicity, you can use four non-overlapping consecutive weeks or twenty two non-overlapping consecutive days to form monthly returns, respectively, if only daily or weekly prices and/or returns are available; students may use whatever software that suits them best and feel free to choose their own data.

Investigations:
The main objective is to use a variety of tests for market efficiency on the above data sets you choose and to write up a concise report on the results of these tests.
Examine, for example,
How well the index returns behavior conforms to a random walk?
Are returns serially correlated?
Is there any difference between the results for the two indices (if you choose two)?
Is there any difference between weekly and monthly frequencies (if you test weekly data)?
Are the results the same over suitable sub-periods (using some economic events that identify
different regimes)? What might the differences (if any) due to?
Testing Statistics:
You are expected to do the following tests to test the weak form efficiency and answer questions outlined above.
First, use Lo and MacKinlays first variance ratio Test Statistics (see Topic 1 slides for detail) to test RW1;
Second, use the heteroskedasticity-robust standardized variance ratio test of Lo and MacKinlay (see page Topic 1 slides for detail) to test RW2 and RW3.