Question regarding Python Code 8.13 (Sharpe Ratio Maximization) in "Advanced Portfolio Optimization"

[mlbhl]

Hello,

I am currently reading your book, "Advanced Portfolio Optimization," and have a question regarding Python code 8.13 (Sharpe ratio maximization).

I'm finding it difficult to understand this specific section of the code and am wondering if there might be an error in the book's content, particularly concerning the line "ret = mu.T @ x". The variable x does not appear to be defined anywhere prior to its use in this line. Could you please clarify if this is intentional or if x should be defined elsewhere?

Additionally, if you are able to answer, I'm curious why the book uses a different approach for defining risk compared to the cp.quad_form() structure commonly used for general optimization with cvxpy. Furthermore, I'd appreciate it if you could explain your reasoning for choosing MOSEK as the solver.

Thank you for your time and assistance.

[dcajasn]

Hi @mlbhl,

I hope you have purchased my book and you are enjoying the content. About your questions:

• It is a typo error "ret = mu.T @ x" must be "ret = mu.T @ y".
• The use of the second order cone formulation of the standard deviation (cp.SOC) is due to accuracy issues with the variance (cp.quad_form), as it is explained in page 127, and because the Sharpe ratio is defined in terms of the standard deviation.
• The explanation of why we use MOSEK is the first paragraph of chapter 7 (page 113).

Best,
Dany

[mlbhl]

Thanks a lot! I’ve been a fan for ages, and I’m excited to dig into the book I just bought.