selecting one of the two report choices:
A. Academic report: Select an academic journal article (approve with me your choice) in the area of Finance on a topic that relates to issues in our class. The article can be new or old so long as it’s a decent article. You will find using Google Scholar, JSTOR, and other online databases helpful for finding the article. Your report should spend a couple of pages summarizing the findings of the paper and explaining how this was a contribution to the field and why it’s interesting. You should then spend a few pages critiquing the paper, identifying shortcomings and possible areas of improvement. You may suggest modifications to the data analysis and/or proofs but you do not need to conduct this work yourself. I will grade you on how thoughtful your analysis, how accurate your summary of the findings are and how much you go above and beyond the bare minimum. Referencing other papers that either support your contest your main paper is perfectly acceptable and would help to improve your grade and bolster your arguments.
B. Industry report: Select an industry which has at least 4 or 5 securities with publicly available data for analysis. These may be stocks or they may be commodities (such as gold, silver, platinum, etc…). The selection of securities should be related however. In the case of stocks all should be from the same industry (such as automotive) or in the case of commodities they should be substitutes or related (such as grains – rice, wheat, etc…). Then identify at least 4 or 5 risk factors and run a multivariate regression of your security returns on the factors you’ve identified. Your report will use the first few pages to summarize the industry you’ve selected and to justify why you believe certain risk factors are pertinent. You may make limited use of summary statistics and/or headline news to make your report more interesting. At a minimum your report should discuss the factor parameter estimates for each security and interpret the sign and significance. You should also discuss the Adjusted R2 of your model, the F-value and any alphas. All of these should be interpreted. Finally, as a minimum you should conduct an out of sample test to see how well your model predicts each security return. You will therefore use time 0 to T data points to estimate your model betas and then use the T+1 data point for your factors and your estimated betas to generate an expected return for T+1 and compare to the genuine T+1 return; state the percent error. How far you go above and beyond the minimum requirements will determine how high a grade you get. Using advanced statistics (ARMA, Durbin-watson, GARCH, etc…), advanced software (SAS, Stata), or clever techniques (Event study, dummy variables, etc…) will all earn a higher grade.
In either case your final report should have all member names and student IDs on the front page and pages should be enumerated. In the case of the academic report I want a link and/or hardcopy of the original paper but in the case of the industry report do not give me the raw data. Simply state where you got the data and the time frame it considers. Reports should be stapled and in the same typeface and font of size 11 or 12 throughout the paper. No additional points will be given for binding, pictures, color graphics unless these are related in some way to a deeper analysis (such as a time-series plot of alpha or the like).