Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Finance 361 Assignment – S2 2023

The deadline for submission is 11:59pm on 22 September 2023 via Canvas

Part A: Portfolio

Background

Today is your first day as an analyst at Alpha Fund, a boutique investment fund located in California which invests in 5 to 15 selected companies. After spending a month training at the New York office, you have some finance knowledge and a smattering of Python coding skills and are ready to get some work done. As you settle in for the day you receive an email from your team manager ……

From: Audrey Koski

To: [You]

Subject: Portfolio

Hello, welcome onboard, I hope everything went well in New York. I’ll be in the office next week Wednesday, then we can have a catch-up. In the meantime, the Investment Committee has asked our team to develop new ideas for investment strategies at the Tuesday meeting. I have narrowed the pool to 20 stocks and would like you to select 7 stocks to form a portfolio. Could you put something together over the weekend? According to the Investment Committee, we are looking for an optimized portfolio that can generate some decent alpha relative to the Fama-French 3-factor model. I have attached a list of specific requirements, please take a look before you start. Your colleague James will show you where to find the data. We also have some basic Python code for you to analyse those stocks. Drop me a memo (5-page max) outlining what you’ve got before Monday!

Good luck,

Audrey

Required

Put together a 5-page memo (in PDF format) addressed to Audrey in which you present the case for your strategy.

Please use the following headings in your memo:

1. Executive summary (100 words max)

2. Stock selection: Pick 7 stocks that could achieve a high Fama-French 3-factor model alpha.  10bp/m is considered high alpha, any higher than  10bp/m will not lead to a higher grade.

3. Preliminary Analysis:

a.   Present  a  table that tabulates the minimum, maximum, mean, median,  and standard deviation of stock returns for each stock.

b.   Briefly explain why you would like each stock to be in your portfolio. One or two sentences per stock should be sufficient.

4. Markowitz optimal portfolio (7 stocks):

a.   Present a correlation matrix of the stock returns of the 7 stocks

b.   Construct  the  Markowitz  optimal  portfolio  for  the  7  stocks  based  on  the historical returns provided

c.   Present the weights allocated to each stock, along with the portfolio’s expected return and Sharpe Ratio (use RF as the risk-free rate – it is included in the dataset).

5. Tikhonov regularisation (7 stocks):

a.   Discuss weights obtained from the Markowitz approach above. How reasonable and implementable are they, in your opinion?

b.   Calculate the Tikhonov optimal portfolio weights (assume lambda equals 0.6) 6. Performance analysis:

a.   Construct  portfolio  returns  using  the  Markowitz   and   Tikhonov  weights. ASSUME that portfolios are rebalanced to match the weights in each period. (A practical side effect of this assumption is that each period portfolio return is simply the weighted average returns of the portfolio stocks in that period.)

b.   For each of the Markowitz and Tikhonov portfolios:

i.   Present the mean and alphas against CAPM, Fama-French 3 factor, and Fama-French 5 factor benchmark models

ii.   Discuss the sign, economic significance, and statistical significance of the factor loadings on the Fama-French 5 risk factors (MktRF, SMB, HML, RMW, and CMA)

iii.   Compare   the  differences  in  portfolio  performance  between  your Markowitz  optimal  portfolio  and  the  portfolio  after  implementing Tikhonov regularisation

7. Recommendation: Recommend one of the two portfolios (Markowitz or Tikhonov) to the Investment Committee. Support your recommendation with the analysis conducted in the main body of the report.

Resources

We have already put together a set of monthly stock return data (from CRSP). In addition we have merged in the monthly factor returns for the 5 factors making up the Fama-French 5-factor model. (More details can be found in Ken French’s data library regarding the dataset:https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html).

In addition, we are providing Python code on Canvas under the folder “Assignment – S2 2023” that produces the required results for the assignment. The code should automatically download and use the data we provide. Some parts of the code are missing, and you will need to complete it (it is part of the assignment!). This will be indicated clearly in the code. See Canvas for more details.

You will need to run Python code to produce the outputs required (e.g., summary tables and  regression  results)  needed  to  complete  the  memo.  You  can  use Google Colab (https://colab.research.google.com/) to run the Python code that we give to you.

While  it's  acceptable  to  utilise  ChatGPT  as  a  helpful  tool  for  your  assignments,  it's important to keep in mind that ChatGPT might not always provide completely accurate information. Therefore, it's advisable to use your own knowledge and critical thinking skills to assess the accuracy of the information provided by ChatGPT.

The data

This is real data (no more playing with pretend data!). The data is provided for educational purposes only; please do not distribute it. The data starts in January  1997 and ends in December 2006 (inclusive) for 20 stocks over 10 years (or 120 months).

Assessment

This part counts for  17% of your  final grade, the other 3% relates to part B of your assignment. The PDF memo must be submitted via Canvas, along with your code. Only the PDF memo will be graded. The deadline for submission is 11:59pm on 22 September 2023. Grades will be awarded as follows:

Stock selection

5%

Preliminary analysis

20%

Discussion of the Markowitz optimal portfolio

20%

Discussion of Tikhonov regularisation

15%

Performance analysis and recommendation

25%

Quality of presentation (professional language, formatting, length etc.)

15%

Good luck!