0
$\begingroup$

I have a research hypothesis and now I'am trying to look at it from different angles.Now I am a bit puzzled.Maybe someone is also interested in machine learning application(especially clustering) in financial area and can advise something.

There are many portfolio theories, for example, Markowitz portfolio theory, that assign weights for every share in portfolio.Imagine, that at the end of a certain year I am estimating results and planning my asset portfolio for the next year. Shares, that perfomed good, can fall during the next year and vice versa. So, it seems reasonable not to rely entirely on the yield and volatility of a security's price over the year.

So, my hypothesis is that we can analyse annual financial reports of stock issuers. For example, choose group of financial multipliers (P/B,P/E,ROE) also maybe compute some annual metrics such as beta coefficient, average price change over the year (in %) and group shares into different segments. After that, we can pick elements from all clusters and only after that assign weights using traditional theories.Ideally, we will get a group of stocks with different performance and seems that it could have a positive impact on portfolio sustainability.

What do you think? Does it sound adequately? Are there any pitfalls in my theory? What to look for when selecting features for clustering?

Thanks everyone in advance!

Improve this question
$\endgroup$
4
  • $\begingroup$ I'm not too familiar with clustering, but using financial ratios/multipliers is a primary function of the Barra risk models, where they perform a cross-sectional regression to deduce the factor risk premia used to decompose the risk of assets. This helps PMs to decide if they want exposure to a certain risk factor. Maybe this will be helpful information to you. $\endgroup$ Commented May 14 at 18:08
  • 1
    $\begingroup$ Of course, with thousands of stocks available to invest, a stock by stock approach is cumbersome. Some sort of grouping or clustering is useful and widely done. There are many different approaches, however, and no unique answer. Google 'hierarchical clustering stocks'. Or start with Building Diversified Portfolios that Outperform Out-of-Sample by M Lopez de Prado 2016. $\endgroup$ Commented May 15 at 7:20
  • $\begingroup$ @nbbo2 thanks. I understand it, I try to combine different fundamental metrics and technical metrics to narrow down the choice with the help of clustering. I want to compare results with Markowitz portfolio with maximized Sharp ratio . Maybe, due to this, it can be possible to outline range of potential features... $\endgroup$ Commented May 15 at 10:02
  • $\begingroup$ Agree with @nbbo2 on this point, I've read that exact work recently, and the paper itself is quite clear and provides support on implementation. I do think you will need some machine learning background though, tree clustering is only the first step in HRP. $\endgroup$ Commented May 15 at 13:15

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.