CAPM is CRAP

some more great stuff

https://www.bloomberg.com/news/articles/2017-04-06/lies-damn-lies-and-financial-statistics?cmpid=socialflow-facebook-markets&utm_content=markets&utm_campaign=socialflow-organic&utm_source=facebook&utm_medium=social

https://blogs.cfainstitute.org/investor/2017/04/05/the-active-equity-renaissance-the-rise-and-fall-of-mpt/?s_cid=smo_CFASocial

his book is on my to read list after drunkards walk

Supposedly, a well diversified portfolio that tends to the market portfolio (say S&P500 for US) has wipeout unsystematic risk. Therefore, the remaining, non-diversifiable risk, is the systematic risk, measured by beta.

The question is… Does beta correctly displays the systematic risk a certain stock bears? If we calculate beta using a linear regression, and assuming the residuals are well-behaved, then beta really do a good job at reflecting the amount of systematic risk the stock bears in relation to the market. If beta is lower than 1, then we know this stock is less risky and should pay less than market, and so on.

The thing is that this method at least gives us some situation where we are.

Active portfolio managers have a lot of work to do to beat index funds if they think active management is as simple as this. Return and Risk analysis in hedge funds should be very different because they are bearing unsystematic risk in order to get active returns. Suitable for wealthy investors tho.

Beta is a garbage tier indicator of risk. It says nothing about the riskiness of a security and you’re lying to yourself if you say otherwise/tell yourself you have some control over “risk” that you don’t in reality, if beta is what you’re using to measure it.

Now to get at the point of whether the average stock picker is better than the market at generating returns and delivering risk adjusted returns… the answer is no.

Risk-adjusted returns are one thing, and returns themselves are other. There are investors that are aware of increased risk exposition, but they seek higher returns (and higher losses if happen). If you want to generate true alpha in a sustainable basis, I will agree with you that it is really really hard, unless you got illegal information (insider).

In my previous post I just wanted to clarify that beta tries to catch systematic risk (macro risk) of a certain security/portfolio. And its usefulness resides in labeling your stock as under/on/above market risk & return with a certain quantitative accuracy. It is not a desvest of returns, nor a sophisticated measure of risk, not even total risk… Perhaps people think beta is the ultimate measure of risk (for even total risk) and get disappointed when do empirical studies or exercises… bad.

thats wrong. there are lot of studies that high beta stocks actually under perform the market and low beta stocks out perform.

its in one of the links

Sorry, couldn’t read all the links. However, my query in short, has those empirical studies showed that in those cases the low beta stocks outperformed the market in the long-term? I mean, if a low beta stock has outperformed the market in a sustainable basis (say 5 years or 10 years) then there is indeed a problem with the model for certain cases. What’s the % of stocks that showed this behavior? If the % is really minimum, say 0.5% of stocks required returns cannot be well predicted by CAPM, then the CAPM is not the problem, right? Those cases are simply outliers. Another question is, any individual event fueled those outperform or under-perform stocks?

I will read the articles.

https://www.analystforum.com/comment/91741753#comment-91741753

I’ve read the article. Well, the problem literally fixed with basic econometrics.

In CAPM defense (LOL), when it was invented, professors had no econometric tools as we do today, so really difficult for the time to find another model that supports reality better.

No wonder why Fama and company found a better fit in the 1993. Computers allowed to do more iterations with relative ease.

APT for the future tho.

I think Beta can be a decent comparative measure as long as it’s not associated with risk.