ASAM Final Report Advice

The structure of your final report should/could follow the following logical outline.

Executive Summary
Report on the Academic Paper(s) that Inspired You
- should describe the motivation, sample period, universe, and alpha of paper(s)
Implementation of Your Actual Strategy
- should include formulas, weighting (signals and investments), universe, sample period, etc. enough to replicate the names you chose.
Backtest Performance
- should include Fama-Macbeth x-sect (avged) regressions, and BJS-FF simple net-of-market, CAPM (1-f), Fama-French (3-f and 5-f); also Sharpe Ratio (plain and annualized). every table should always include information like N and time-period, frequency and other details necessary to replicate your investment weights.
Actual Performance
- should include inception day, inception weights, cash holdings, beta (and how calculated...ex-ante vs ex-post? daily?), Sharpe ratio. also an industry table (perhaps with performance of industry over sample period vs. your stocks in industry) and a table with best and worst performers (with investment weights).
- it should also include a performance attribution. that is, run a factor time-series regression on your daily realized rates of return (i.e., rp--rf is your dependent variable). the mean of your dependent variable is by mathematical necessity equal to your alpha plus the beta1 times the mean of factor1, etc. The "estimate beta1 times mean of factor1" is what part of your portfolio excess return can be attributed to your exposure on factor1. also, do the same kinds of attribution regression industry by industry, always including the market premium, though. (it helps telling you whether your performance was due to your industry exposure or due to your stock selection.)
- at the end, you can add the customary bull shit that holds forth about why your xyz holding did so poorly (it's the Fed! no, it's profit-taking! no, it's tea leaves! no, it's ... . in some cases, you can actually determine the cause [e.g., a firm reported terrible earnings and its drug was rejected by the FDA], in other cases, it's pure speculation [e.g., investors worldwide started to hate solar tech]. when you describe your BS, explain that you are guessing and make sure that it is logical and not just hot air.)
For clarification, we do this not because we believe that a few months of actual performance is representative of any strategy, but because ASAM is primarily a learning experience.
Thoughts on Future Changes
- what you would have done if you had had another 6 months (or were really thinking of exploring before running the strategy in a real fund).
Appendix: Alternatives considered. Problems. Etc. Portfolio changeover. Some sample calculations on the formulas/weights you used---perhaps for every stock in your incepted portfolio.
Appendix: Fully documented source code with instructions. Replicable.

other observations: always be clear about what statistics you are displaying. is excess return rp-rf or rp-rm? always be clear about units (is 0.09 alpha 9bp/day or 9%/year or ...?). 3000bp is a bit silly, say 30 points or 30%. use a reasonable number of digits for precision. use reasonable table justification. use a minus sign instead of a hyphen for negative numbers. don't use 3 point fonts. avoid empty useless filler phrases. stick to one tense. don't try to be fancy on colors, headlines, etc. (it's not a grade school report.)


## Order of Presentation

Reports should be written newspaper style --- i.e., in order of importance, not in chronological order.

Think about it --- who has time to wade through pages of information only to arrive at the meat of what you were doing by the time they are asleep?

This usually means that it starts with an executive summary, then describes the performance, and only then all the finer details.


* Executive Summary: typically 1 sentence of the strategy idea basics, 1 sentence performance metrics, 1 sentence whatever.

* Performance Details: typically cumulative plot (also benchmarks), Sharpe Ratio, alphas.

* Strategy Details

  - Motivation.
  
  - Description (incl. weighting, example calculation).  If strategy is intricate, then summarize only and refer to a detailed appendix.

  - Backtest Performance, published.
  
  - Backtest Performance, your own tests (e.g., use annual buy-and-hold etc.)

* Performance and Attribution

  - Inception

    - Sector Allocations, Biggest Holdings, etc.

    - Exposures

  - Experienced Performance
  
    - various benchmark and factor models

* Suggestions for Future Cohorts

* Would you put your own money into this strategy now?


* Appendix 1: Details of Strategy

* Appendix 2: Only for Anderson (grade): what you tried that did not work.

---

Q: A few groups stated that VTV was a proxy for HML.  Is it?  How would you test this?  How would you find out how you could proxy better?  Would you want to proxy for HML?

units on each table and figure (annual?, daily?, percent?), sample period, variable definitions. performance --- one digit after decimal point on percent per year. ex: "3.4%" grammar and spell errors are a sign of carelessness and little consideration for the reader. after all, you have 4-5 people writing a report. can't one of them read it afresh? since -> because when logical. since when time precedence is meant. avoid passive tense. don't break tables unless absolutely necessary. is every sentence in its correct logical spot, or are you randomly inserting sentences? Additional work --- "if we could do it over again for a "version 2," ..." ---------------------------------------------------------------- performance attribution as, while, because, since, although, whereas few adverbs and extreme adjectives, please.