Math Pedantic/Computer Science Honours interested in Cause Prioritization. Currently working with QURI on Squiggle and Pedant.
And to add to this, very recently there was a post Quantifying the Uncertainty in AMF! Which still seems a bit in the works but I'm super excited for it.
My Hecking Goodness! This is the coolest thing I have ever seen in a long time! You've done a great job! I am like literally popping with excitement and joy. There's a lot you can do once you've got this!
I'll have to go through the model with a finer comb (and look through Nuno's recommendations) and probably contribute a few changes, but I'm glad you got so much utility out of using Squiggle! I've got a couple of ideas on how to manage the multiple demographics problem, but honestly I'd love to have some chats with you about next steps for these models.
Hello! My goodness I love this! You've really written this in a super accessible way!
Some citations: I have previously Quantified the Uncertainty in the GiveDirectly CEA (using Squiggle). I believe the Happier Lives Institute has done the same thing, as did cole_haus who didn't do an analysis but built a framework for uncertainty analysis (much like I think you did). I just posted a simple example of calculating the Value of Information on GiveWell models. There's a question about why GiveWell doesn't quantity uncertainty
My partner Hannah currently has a grant where she's working on quantifying the uncertainty of other GiveWell charities using techniques similar to mine, starting with New Incentives. Hopefully, we'll have fruit to show for other GiveWell charities! There is a lot of interest in this type of work.
I'd love to chat with you (or anyone else interested in Uncertainty Quantification) about current methods and how we can improve them. You can book me on calendly. I'd still learning a lot about how to do this sort of thing properly, and am mainly learning by trying, so I would love to have a chat about ways to improve.
Hello! Thanks for showing interest in my post.
First of all, I don't represent GiveWell or anyone else but myself, so all of this is more or less speculation.
My best guess as why GiveWell does not quantify uncertainty in their estimates is because the technology to do this is still somewhat primitive. The most mature candidate I see is Causal, but even then it's difficult to identify how one might do something like have multiple parallel analyses of the same program but in different countries. GiveWell has a lot of requirements that their host plaftorm needs t ohave. Google Sheets has the benefit that it can be used, understood, and edited by anyone. I'm currently working on Squiggle with QURI to make sweeten the deal to quantifying uncertainty explicitly, but there's a long way to go before it becomes somehing that could be readily understood and trusted to be stable like Google Sheets.
On a second note, I would also say that providing lower and upper estimates for cost-effectiveness for its top charities wouldn't actually be that valuable, in the sense that it doesn't influence any real world decisions. I know that I decided to spend hours making the GiveDirectly quantification but in truth, the information gained from it directly is extremely little. The main reason I did it is that it makes a great proof of concept for usage in non-GiveWell fields which need it much more.
There are two reasons why there is so little information gained from it:
I see much more value in quantifying uncertainty when we might expect the uncertainty to be much larger, for instance, when dealing with moral uncertainty, or animal welfare/longtermist interventions.
Haha, I came up with that example as well. You're thinking about this in the same way I did!
I think to say that one is the "actual objective" is not very rigorous. Although I'm saying this from a place of making that same argument. It does answer a valid question of "how much money should one donate to get an expected 1 unit of good" (which is also really easy to communicate, dollars per life saved is much easier to talk about than lives saved per dollar). I've been thinking about it for a while and put a comment under Edo Arad's one.
As for the second point about simple going E(cost)E(effect). I agree that this is likely an error, and you have a good counterexample.
Thank you so much for the post! I might communicate it as:
People are asking the question "How much money do you have to donate to get an expected value of 1 unit of good" Which could be formulated as:E(good(x))=1
where x is the amount you donate and good(x) is the amount of utility you get out of it.
In most cases, this is linear, so: good(x)=goodcost∗x. And E(goodcostx)=1.
Solving for x in this case gets x=E(goodcost)−1, but the mistake is to solve it and get x=E(costgood).
Please correct me if this is a bad way to formulate the problem! Can't wait to see your future work as well
That's true! η could easily be something other than 1.5. In London, it was found to be 1.5, in 20 OECD countries, it was found to be about 1.4. James Snowden assumes 1.59.
I could but don't represent eta with actual uncertainty! This could be an improvement.
Now that I've realised this, I will remove the entire baseline consumption consideration. As projecting forward I assume GiveDirectly will just get better at selecting poor households to counteract the fact that they should be richer. Thanks for pointing this out!
Oh no, I've missed this consideration! I'll definitely fix this as soon as possible.
Would love to! I'm in communication to set up an EA Funds grant to continue building these for other GiveWell charities. I'd also like to do this with ACE! but I'll need to communicate with them about it.