Note that in the context of trading/investing, the two terms are often used differently. There, “mean reversion” often means negative autocorrelation of returns, which can either be ~causal or driven by price level noise (which in turn is more like a “regression to the mean” idea). If you invest in a mean reversion strategy you tend to have an actual mechanism in mind though.  

“Regression to the mean” is a less ambiguous  term and generally means what you describe. 

Thanks a lot Joey, this is definitely worth reading for people in the wider EA space, not only larger scale donors or people working in philanthropy directly. 

What I’ve found particularly helpful are the rough quantitative guidelines regarding “charity time consumed per amount donated” and “how to donate as a function of annual amount and time spent per week”. 
 
This is very valuable to better position myself from an earning-to-give perspective. 

I think it might perhaps be interesting to write a short summary of that for the forum, perhaps targeted more at a median e2g EA? (If that doesn’t exist already.)

Separately, it’s great to see that the book really embraces plurality in what areas donors prioritise without too strong a view on what’s preferable in the author’s opinion. 

Yes thanks 

boy did this age in favour of "good judgement" as a factor!

To add a small side note to this, in particular the point around the effectiveness essay: 

I suspect the EA community and in particular 80k hours tend to underestimate how hard it is to do better by being more ambitious (for the typical engaged EA, at least). Eg counterfactually increasing your income from 150k to 600k by "being more ambitious" and working longer hours or negotiating your salary more aggressiely is not a very high probability outcome. Achieving this increase by having better judgement around what area to specialise in is perhaps more likely. Likewise, taking more risk by being an entrepreneur does not 10x your career donations in expectation if you have a decent job.
I would discount the multipliers in 6 & 8 a lot (or at least their component attributable to ambition and risk taking), while I believe they are > 1.

Just to point out the obvious: encouraging some of these professionals to think more about earning to give can also be very valuable.

That’s right, but it should be possible to model that in a very similar hierarchical manner and adjust accordingly, too, if you buy into the original framework laid out in the post.

(I haven’t fully thought it thru but it does strike me as fundamentally possible with the same caveats of not knowing parameters, not that I’d suggest using the toy model style maths in practice). 

Thanks for sharing this, awareness of this type of bias is very relevant for the EA community. 

The interpretation of $\sigma_V / \sigma_\mu$ (squared) is subtle in practice. I think a clean way to express it is the (square root of the) ratio of prior precision to “measurement” precision - that fits with the hierarchical model used to explain it in the paper you reference.

In practice this is not trivial to guesstimate. 

An interesting rabbit hole to understand this further is the “Tweedie correction” [1]. 

It should also be pointed out that once you’ve shrunk the estimate, that’s it: EV maximising will pick the posterior winner without accounting for the posterior variance - also something not everyone is comfortable with. 

[1] https://efron.ckirby.su.domains/papers/2011TweediesFormula.pdf

“Trump is pressuring the Fed to adopt policies that would cause inflation.”

That’s more cleanly expressed as a curve steepener (front lower, back higher), so bullish short end vs bearish back. 

“AI-induced job loss might cause the Fed to be less concerned about inflation.”

This sounds more bullish bonds because low inflation concerns -> fed can cut. Also (more importantly) the fed has a dual mandate so low employment -> cut.