Prof. Prem Jain

1. Understanding of Financial Accounting Research published in such journals as The Accounting Review, Journal of Accounting Research, and Journal of Accounting and Economics.
2. Conducting Financial Accounting Research -- a research paper is required. You should soon (within the first two weeks from the first day of classes) give me a proposal of the research that you are going to conduct.

1. Statistics -- Hypothesis testing and regression analysis. We will review some of this material, if needed.
2. Basic Economics
3. Basic Finance -- We will review the basic portfolio theory, if needed.

Financial Reporting: An Accounting Revolution, Third Edition, by William Beaver, Prentice Hall.

"Statistics for Business and Economics" by Anderson, Sweeney and Williams

Please review Simple Linear Regression (Chap. 14) and Multiple Linear Regression (Chap. 15) before coming to the class. You can study this material from anywhere you like but if you do not have a statistics textbook, please read chapters 14 (Simple Linear Regression) and 15 (Multiple Linear Regression) on these topics that I have placed in library reserve. These chapters are from Statistics for Business and Economics by Anderson, Sweeney, and Williams (sixth edition).

Tentative Schedule: You would have learned a lot if we can cover the following.

Aug. 31(Tue): Introduction. (Assignments 1 and 2 given to students)

Sep. 2 (Thu): Simple Linear Regression (Chapter 14 from the statistics book)

Sep. 7 (Tue): Multiple Linear Regression (Chapter 15 from the statistics book)

Sep. 9 (Thu): HW No. 2 (a little project on Multiple regressions) is due. HW No.3 and 4 (event study, individual basis and group project explained)

Sep 10 (Fri): Continuation of the Sep. 9 class.

Sep. 14 (Tue): Event Studies: How are they conducted, statistical tests, etc.,

Sep. 16 (Thu): Discussion of Beaver's book, especially Ch. 5 and Ch. 6

Sep. 21(Tue): Discussion of Beaver's book (continuation) and student projects.

Sep. 23(Thu): Students to discuss their projects in the class to get feedback from everyone.

Sep. 28(Thu): Journal article (Research paper) No. I (Exact titles to be announced, see note below)

Sep. 30 (Thu): Journal article No. I

Oct. 5 (Tue): End Journal article I and start Journal article II.

Oct. 7 (Thu): Same as above. During all this period, also discuss how students are progressing with their research projects.

Oct. 12 (Tue): Same as above.

I think that the second course (Financial Accounting Seminar II) will include more research papers and additional research methodologies (statistics), etc. Hopefully, the material covered here will help you improve your thinking about research in accounting and finance.

There will be a few small unannounced tests. There is no final test but in class participation during the discussion is important. The final project is due on the day you promise delivery.

Note: Most probably, the two journal articles are (unless I find more interesting ones):

1. "An Analysis of the Recommendations of the "Superstar" Money Managers at Barron's Annual Roundtable." by Hemang Desai and Prem C. Jain, Journal of Finance, September 1995, pp. 1257-1273.
2. "Evidence that Stock Prices do not Fully Reflect the Implications of Current Earnings for Future Earnings." by Victor Bernard and Jacob Thomas, Journal of Accounting and Economics, 1990, pp. 305-340.

Statistics book, Chap. 14, problems number 5, 18, and 26 (Optional: 35).

(This is similar to the in-class problem (answers distributed): Chap 14, Problem no. 3, 17, 25, and 34 (Prob. no. 34 was not really discussed).

1. For the thirty Dow Jones stocks, find the following:

1. Price per share (pick a date, say today).
2. Earnings per share -- latest available
3. Predicted earnings per share, i.e., analysts' forecast of eps for the current year, for the year after the current year, and expected growth for the next five years.
4. Book-value per share.

1. Estimate a number of regressions explaining price per share.

Write a small report (half a page, one page, etc.) Your creativity is in (e) and in your interpretation of the results. Please do not spend much time on making the report look nice. Your time is better spent on thinking about the issues involved.

The purpose of this exercise is to get a better understanding of the effect of earnings on stock prices. Think of four companies, preferably two small and two large. Now think of a year between 1991 (Year one) and 1996 (Year 6), chosen randomly. To randomize this, use the year that corresponds to the month of your birthday. If you were born in January or February, select year 1991 and so on. Find the earnings per share for the four companies that year and compare it with the earnings per share in the year earlier. To get this data, you might like to use Bloomberg or some other source. These two yearly numbers should be comparable in the sense that if the company had a large merger, etc., do not include that company in your list. Of course, if you do not know that or if it is a small merger (or some other event), the firm may be included. In general, diluted eps will suffice. But in some cases, those numbers are not reliable or meaningful. I recommend that if in doubt you may like to use earnings before X.O. items and discontinued operations or operating earnings, etc. The basic idea is to find out how well (in percentage terms) have the companies done relative to the previous year. For example, let us say that you compare the eps of KO (Coca Cola) in the two years and conclude that KO's operating earnings went up by 7%. Similarly, you find this (% change) for the other three companies. I would like you think about this and pick companies that you know a little bit about but do not try to pick only those companies that have done extremely well. We should be unbiased as far as possible. Not all companies do well every year.

For all the four companies, find monthly returns for the two-year period from the year before to the year after. Assume that you selected WMT (Wal-Mart) and the year is 1993. Wal-Mart's fiscal year ends in January. Thus, you should get monthly returns from February 1992 to January 1994 (24 monthly returns). Year 1: First twelve monthly returns (sum of twelve observations), Year 2: Second twelve monthly returns (sum of twelve monthly observations).

Finally, find the corresponding monthly returns for the market. There are many sources for this. I can send you an Excel worksheet if you want. In some sense, we are trying to understand relationship between earnings and stock prices and some of these ideas are explained in Chap. 5 of Beaver's book. The core of this idea can be found in Ball and Brown (1968, Journal of Accounting Research). This will also form the basis of many other research articles in accounting and finance, including the articles we are planning to do in Seminar I (and II). Hopefully, this will make you think about these issues in an intelligent manner.

Although you have only four data points (four firms), do you see the following:

1. For the first twelve months, is there a relationship between percentage change in earnings and raw returns (i.e., for the entire Year 1). You do not need to estimate any regression, etc. Simply by looking at the numbers, can you say something?
2. : ......... abnormal returns (firm's returns - market return).
3. Same as (1) for the second year.

(4) Same as (2) for the second year.

You can find monthly returns on Bloomberg. Bloomberg definitely has stock prices (already adjusted for splits, etc.). For this exercise, you can ignore adjustment for dividends.