Status Updates From Advanced Portfolio Manageme...

Wait it turns out that this Rolf Banz guy discovered the size factor before Fama & French. No wonder Fama keeps saying that the evidence was smacking him in the face already, and it was just like stabbing in a barrel full of fish.

What's fascinating is that it turns out that apparently, modern research considers the size effect to not be a real effect?! Wtf is a 'real effect'? Gappy cites Alquist et al (2018) (JPM).

Note 2/2:
... the 1st place lmao.

Not to mention that, take the 3-factor model for example, Fama and French had clear definitions to sort for HML and SMB.

Note 1/2:
A key point that I don't understand — Gappy teaches that for CAPM, we can obviously observe the market factor (since it's just the ERP), but we cannot observe the auxiliary factors. I don't get it — take the profitability factor like Gappy does: don't we observe the P/CF of the cross-section of all stocks? If you can't even observe the factors, then how tf did the researchers build factor portfolios in ...

From the discussion thus far, Gappy explains that principals (investors in the fund) can see the alpha-beta decomposition. I wonder if there's some software for LPs to easily/clearly see the decomposition/performance attribution. The math certainly isn't difficult, it's just basic spreadsheet calculations — even the CFA teaches it. So it should be easy to commercialize in some software.

Note 2/2:

Recall that log-normal distributions are used to model stock prices, asset returns and exchange rates cuz they tend to be positive and right-skewed. The log scale normalizes a positive and right-skewed distribution.

Note 1/2:
I don't understand why Gappy straight-up claims that the daily returns of a single asset 'are not even log-normally distributed'. It's reasonable to say that they are 'not usually log-normally distributed', but to straight up say that they are NOT seems strange (but intellectually humble at the same time I suppose).

Gappy claims that, 'if enough care is put into the models (for equities), the factor and idio returns are sufficiently well-behaved, ... **higher-order stats of returns, e.g. skewness and kurtosis, cannot be estimated**'.

Isn't this just factually incorrect? I'm pretty sure there are plenty of papers estimating the skewness and kurtosis of equity returns.

Note 2/2:
... the sell-side for management access), and even prices in their views!? What?!

Note 1/2:
'Sell-side upgrades/downgrades are an important source of revenue for fundamental PMs' — I never really understood this. It’s intuitively (and casually perhaps) observed that sell-side upgrades/downgrades really do move stock prices at times. So, the market (or the buy-side) actually *does* give a shit about the sell-side’s price targets then (instead of the common adage that the buy-side only uses ...

Note 2/2:
... won’t continue to pay them for much longer), but Gappy makes an interesting point here — beta (i.e. macro risk basically) is unknowable (almost by definition), so we should hedge out what we don’t know, to limit our unintended bets. A view that I must think about for some more time, and I wonder if the typical pod shop analyst holds this view.

Note 1/3:
From the way this is leading, this book looks like it's gonna be about the typical pod-shop obsession to isolate stock-specific alpha, and hedge away all the beta. I never really understood the obsession to NOT earn any beta, since beta are an absolute source of return as well, other than from a professional perspective (cuz obviously, professional PMs cannot beta-ride or else their clients presumably ...