Compute the daily price spread of the asset, defined as the daily high minus low price. Choose an appropriate model, an in-sample period and an out-of-sample period, and conduct a rolling window forecasting exercise for the one-day-ahead spread. Evaluate the forecasting performance of your model and compare it with a simple AR(1) model using loss functions and statistical tests. Interpret your results in light of the stylized facts of return volatility.

In-sample and out-of-sample data is provided in Excel spreadsheets.

Your report is expected to include the following contents: 1. A short introduction to the problem describing the financial meaning and the importance of the problem from a financial point of view in your own words. 2. A short description of chosen econometric methodology with appropriate academic references to the relevant textbooks or papers. You are not required to present mathematical formulas (unless they are necessary for presentation or explanation purposes), but you should clearly state the hypotheses, decision rules and implications of all hypothesis tests. The details of the implementations should be clearly described as well. 3. The results of the empirical analysis in the form of figures and tables. Figure/table should be properly numbered and each of them should include a short and clear caption. 4. Interpretation and discussion of the empirical results (including the discussion of limitations if any). 5. The empirical analyses should be performed in MATLAB. You should include all the written MATLAB codes used to perform the empirical analysis as an appendix, attached to the end of the report.

Length 2.5 pages including tables and graphs.



Rolling window forecast


AR model

Powered by WordPress