Assignment 3: Case: AQR’s Momentum Funds (A)
Due Sunday by 11:59pm Points 10 Submitting a file upload Available until Jun 13 at 5pm
Case: AQR’s Momentum Funds (A). The case can be obtained at https://hbsp.harvard.edu/import/906634
Submit a report answering the following questions in three UNCOMPRESSED files: 1). A report as a Jupyter notebook (extension ipynb) organized by questions, with the proper explanation for every answer (50% penalty for missing this file), 2). the same Jupyter notebook saved as an HTML or pdf file (it might be easier to generate an HTML file than a pdf file; 20% penalty for missing this file), and 3) The data used as input in your program (10% penalty for missing this file).
Copyright By PowCoder代写 加微信 powcoder
Questions:
1. Should AQR launch its momentum funds?
2. Do you believe the Fama-French momentum (UMD/MOM) factor will have returns over the next decade that are significantly greater than
zero, significantly less than zero, or approximately zero? Use the historical data to do your analysis.
3. Compare the UMD factor to other specifications for momentum. Specifically, use the Spreadsheet Supplement for the case (which
contains the momentum decile returns from Exhibit 4 along with the time series of the UMD return) to create the following four momentum portfolios and generate the average returns for each of them for every decade (the 1920s, 1930s, etc.):
1. Decile Spread portfolio returns = (10-1)
2. Quintile Spread portfolio returns = ((10+9)-(1+2))
3. UMD spread portfolio returns (given in Spreadsheet Supplement) 4. Median Spread portfolio returns = (10+…+6)-(1+…+5))
5. Does this affect your answer to (2) above?
4. What are the appropriate benchmarks for AQR’s Momentum Funds? Will the net performance of the funds exceed those benchmarks? Why or why not?
5. The advantageous correlation structure in Exhibit 5 was seen as a key selling point of momentum: 1. Is this the right way to think about AQR’s Momentum mutual funds?
2. If not, use the data in the Spreadsheet Supplement to construct a more informative set of correlations. Does anything in AQR’s story change?
6.Does momentum make an attractive product for retail mutual fund investors?
7. If AQR launches its momentum funds, how should the firm weigh maximizing returns vs. minimizing tracking error? How should it manage the portfolio?
8. Build a multifactor model using the period 1927-1999 to forecast the annual return of your preferred stock for the period 2000-2008 based on the four specifications of the UMD factor from question 3, and the three original factors: Mkt – RF, HML, and SMB with the following methodologies and calculate the mean squared error (MSE):
1. Linear regression
2. Ridge regression
3. Lasso regression
4. Random Forest for regression 5. Decision tree for regression 6. Support vector regression
7. Build a table with the MSE of the above algorithms and discuss your results. —————-
Please include the following pledge at the top of your solution:
I affirm that I will not plagiarize, use unauthorized materials, or give or receive illegitimate help on assignments, papers, or examinations. I will also uphold equity and honesty in the evaluation of my work and the work of others. I do so to sustain a community built around this Code of Honor.
Signature _________________________ Date: _____________
Course Chat
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com