Topic · Systematic Strategies & Alpha
← All topicsThe Factor Zoo
30 articles · 4 checkpoints · 18 deeper reads · 8 reference notes
A standalone topic: it is on no roadmap, so read it on its own terms.
Every article, in reading order
plant a flag as you finish eachRead these first
Campbell Harvey, Yan Liu and Heqing Zhu counted hundreds of published factors and asked the obvious uncomfortable question: if you test that many ideas, how many would look significant by pure chance? Their answer was that the standard significance bar finance uses is far too low, and most "discovered" factors should not have cleared it.
Kewei Hou, Chen Xue and Lu Zhang took hundreds of published stock-return anomalies and re-ran every one with the exact data and methods the original papers used. Roughly half failed to replicate at standard significance, and the paper became the single most cited stress test of the factor literature.
The same factor idea, value, say, can look strong or weak depending on which price ratio you use, how you weight the portfolio, how often you rebalance, and which stocks you exclude. None of these choices is wrong on its own, but stacked together they can flip a factor from significant to noise, which is why two "value factor" papers can disagree.
R. David McLean and Jeffrey Pontiff tracked what happened to 97 published stock-return anomalies after their papers came out and found something clean: returns fell by about a third after publication, exactly as you'd expect if the paper itself was what killed the trade.
Then the rest