1) Introduction: e.g., where does the idea come from (new or based on existing research), what...

Transcribed Image Text:

1) Introduction: e.g., where does the idea come from (new or based on existing research), what is the motivation/general idea/mechanism, a quick summary of your findings. Proper citations are needed (the bibliography should be included at the end of the report). 2) Data: description of the data source, data cleaning/filtering process, and sample period. 3) Empirical Analysis: You will describe the trading rules (how you execute the strategy) in details; report the performance of your strategy (backtesting); and risk management methods (optional). Just to give you some ideas: a) For trading rules: how do you rebalance your portfolio, how frequently do you rebalance your portfolio, how do you weight each security and determine its position. b) For performance measurement: what are the relevant benchmarks/factor controls, significance of alphas, performance during different sub-samples, impact of transaction costs (one way to study it is to use different rebalance frequency), out-of-sample tests (if working on a market-timing strategy), impact of using different weights for securities in a portfolio. c) For risk management: historical drawdown; potential drawdown control or VaR. Also, figures are as important as tables. For example, you can visualize the performance by showing the strategy's cumulative returns alongside other benchmarks (what I did in Lecture 1), or show the history of drawdown. Finally, it is helpful if you can compare your results to the existing work in the finance academic literature. 4) Conclusion and Discussion: A summary of your results. Also a discussion of why your strategy works/does not work so well. Suggestive Topics: 1. Test the performance of anomalies (utilizing the "Value" code from Canvas) by choosing your own signal (e.g., quality, profitability, investment...). Replicating an existing academic paper with extended sample period/different markets is also acceptable. 2. Test the predictive power of predictors for aggregate market returns. For example, is price-dividend ratio in Hong Kong a strong predictor of Hang Seng Index excess returns? What is its in-sample and out-of-sample performance? 3. Predictability of currency returns. Are currency carry trade returns predictable by some macro/financial variables? 4. Any strategy covered in the class but not implemented in R, e.g., Treasuries yield curve trades. 1) Introduction: e.g., where does the idea come from (new or based on existing research), what is the motivation/general idea/mechanism, a quick summary of your findings. Proper citations are needed (the bibliography should be included at the end of the report). 2) Data: description of the data source, data cleaning/filtering process, and sample period. 3) Empirical Analysis: You will describe the trading rules (how you execute the strategy) in details; report the performance of your strategy (backtesting); and risk management methods (optional). Just to give you some ideas: a) For trading rules: how do you rebalance your portfolio, how frequently do you rebalance your portfolio, how do you weight each security and determine its position. b) For performance measurement: what are the relevant benchmarks/factor controls, significance of alphas, performance during different sub-samples, impact of transaction costs (one way to study it is to use different rebalance frequency), out-of-sample tests (if working on a market-timing strategy), impact of using different weights for securities in a portfolio. c) For risk management: historical drawdown; potential drawdown control or VaR. Also, figures are as important as tables. For example, you can visualize the performance by showing the strategy's cumulative returns alongside other benchmarks (what I did in Lecture 1), or show the history of drawdown. Finally, it is helpful if you can compare your results to the existing work in the finance academic literature. 4) Conclusion and Discussion: A summary of your results. Also a discussion of why your strategy works/does not work so well. Suggestive Topics: 1. Test the performance of anomalies (utilizing the "Value" code from Canvas) by choosing your own signal (e.g., quality, profitability, investment...). Replicating an existing academic paper with extended sample period/different markets is also acceptable. 2. Test the predictive power of predictors for aggregate market returns. For example, is price-dividend ratio in Hong Kong a strong predictor of Hang Seng Index excess returns? What is its in-sample and out-of-sample performance? 3. Predictability of currency returns. Are currency carry trade returns predictable by some macro/financial variables? 4. Any strategy covered in the class but not implemented in R, e.g., Treasuries yield curve trades.