To get this to look right, you'll need to import the stylesheet as well and run it. If you'd like I can patch up the notebook

Might be good to introduce and explain Squiggle as well

I am now importing the stylesheet, but I don't understand what you mean by "run it". If you can help patch up the notebook, that would be great. I didn't copy directly from your other notebooks because I want to avoid scaring readers with code they don't understand. I am therefore trying to "hide" the technicalities in the notebooks I am important from.

I have added a brief explanation of Squiggle with a pointer to your tutorial.

Do you mind if you can cute why StrongMinds is presumably the most cost effective? As well as all the other cost effectiveness figures?

To me, these two questions are the same.

There is a subtle difference between the two questions because the amount of good we can do additionally depends on how much money will be invested.

I have rewritten the text to avoid this confusion.

I understand this math, but might be good to cite Wikipedia or something when talking about them. Might be a bit confusing for others

I am already citing Howard and your EA Forum post. Did you see anything in my equations that is not covered by those sources? I am accustomed to academic writing where citing Wikipedia is not an option.

If it helps, I can also write the equations in the text.

Technically, the expected change in ce is the value of perfect information. This figure seems to represent the amount of extra good we can do with a given budget with the information. (If the information cost was 0)

As far as I can tell, the definitions given by Howard and Wikipedia do NOT normalize the decision's outcome by the cost of the decision.

You are right that these calculations assume that the cost of information is zero and I will acknowledge that.

In addition, I have clarified that my calculation implements the definition given by Howard (1966).

There's a better syntax for multiplets in the new update

Where can I read about or see an example of it? If you like, you can illustrate it by editing this cell.

Heitjan et al. (1999) argue that the vector (effectiveness, cost) can be assumed to follow a bivariate normal distribution. Under this assumption, the cost-effectiveness would follow the ratio distribution derived by Hinkley (1969). https://stats.stackexchange.com/questions/202179/what-is-the-distribution-of-the-ratio-of-two-normals Hinkley D.V., 1969, On the Ratio of Two Correlated Normal Random Variables, Biometrica vol. 56 no. 3