Tuesdays are all about academic (and practitioner) literature at Abnormal Returns. You can check out last week’s links including a look at the rise and fall of mean reversion as a factor.
Quote of the Day
"(S)upervised learning" -- the main strand of machine learning -- is just function estimation, and in particular, conditional mean estimation. That is, regression. It may involve high dimensions, non-linearities, binary variables, etc., but at the end of the day it's still just regression."
(Francis Diebold)
Factors
- An in-depth look at the basis for the value and momentum effects. (osam.com)
- Why fixed income investors shouldn't ignore style factors. (academicinsightsoninvesting.com)
- How to think about two major issues with factor regressions. (alphaarchitect.com)
- What makes for stable, sustainable factor returns? (etf.com)
- Do Fama-French factors (size and B/M) replicate in China? (academicinsightsoninvesting.com)
Backtests
- Why it is so easy to get fooled by a backtest. (alphaarchitect.com)
- Summary backtest statistics can hide a whole lot behind the scenes. (alphaarchitect.com)
- Be wary of high CAGRs: the starting date matters. (priceactionlab.com)
Research
- Under what conditions, i.e. trend, does dollar-cost averaging make sense? (blog.thinknewfound.com)
- Why hedge fund returns should NOT be compared to the S&P 500 or any other stock market index. (aqr.com)
- Good luck trying to use CAPE ratios to time the stock market. (factorresearch.com)
- Today's accounting systems are broken when it comes to intangibles. (bloomberg.com)
- There is evidence of seasonality in sector returns. The bigger question is why? (blog.thinknewfound.com)
- Risk premia strategies have come to the commodities futures space. (ft.com)