Vasco Grilo🔸 comments on Uncertainty over time and Bayesian updating

[Vasco Grilo🔸]

Great post, David! Probably my favourite of the series so far.

With these assumptions in place, we can derive the agent’s posterior expected value of the intervention at each time period (see Proposition 1 in Gabaix and Laibson (2022) and proof of Proposition 1 in Appendix A).

$E\left(u\right(t\left)|s\right(t\left)\right)=D\left(t\right)s\left(t\right)=\frac{s\left(t\right)}{1+\frac{f\left(t\right)}{{\sigma }_u^2}}$

$D\left(t\right)\equiv \frac{1}{1+\frac{f\left(t\right)}{{\sigma }_u^2}}$

$D\left(t\right)$ captures how much weight the agent puts on the signal. Alternatively, it captures how much the agent ‘discounts’ the signal solely due to the uncertainty surrounding the signal. In the specific case where the prior is mean 0 and the signal is mean 1, $D\left(t\right)$ represents the posterior expected value.

One can think about this along the lines of the inverse-variance weighting. The formula above is equivalent to “expected posterior” = “signal precision”/​(“prior precision” + “signal precision”)*”signal”, where “precision” = 1/​”variance”. This is a particular case (null prior) of Dario Amodei’s conclusion that “expected posterior” = (“prior precision”*”prior” + “signal precision”*”signal”)/​(“prior precision” + “signal precision”). As the precision of the signal:

• Increases, the expected posterior tends to the signal.

• Decreases, the expected posterior tends to the prior.

This gives us a mean error of −0.054 as shown in Table 2. In other words, on average the surrogate index approach underestimates the benefits of treatment by 0.047 standard deviations.

-0.054 is supposed to be −0.047.

Constant variance prior: We assume that the variance of the prior was the same for each time horizon whereas the variance of the signal increases with time horizon for simplicity.

[...]

If the variance of the prior grows at the same speed as the variance of the signal then the expected value of the posterior will not change with time horizon.

I think the rate of increase of the variance of the prior is a crucial consideration. Intuitively, I would say the variance of the prior grows at the same speed as the variance of the signal, in which case the signal would not be discounted.

Normal distributions: We use normal distributions for mathematical convenience but non-normal distributions with fat tails are everywhere in the real world.

If the signal and prior follow a lognormal distribution, their logarithms will follow normal distributions, so I think one can interpret the results of your analysis as applying to the logarithm of the posterior.

[David Rhys Bernard]

Thanks Vasco, I’m glad you enjoyed it! I corrected the typo and your points about inverse-variance weighting and lognormal distributions are well-taken.

I agree that doing more work to specify what our priors should be in this sort of situation is valuable although I’m unsure if it rises to the level of a crucial consideration. Our ability to predict long-run effects has been an important crux for me hence the work I’ve been doing on it, but in general, it seems to be more of an important consideration for people who lean neartermist than those who lean longtermist.