Issue: Chapter 10, very minor typo

[TimoFlesch]

hey, first of all I'd like to thank you for providing such a fantastic resource! Apologies for being pedantic, there is a very minor issue on chapter 10, in the following two paragraphs:

This estimate is done by 1) partitioning the data into confounder cells, in this case, man and women, 2) estimating the effect on each cell and 3) combining the estimate with a weighted average, where the weight is the sample size of the cell or covariate group. If we had exactly the same size of man and woman in the data, the ATE estimate would be right in the middle of the ATE of the 2 groups, -2.5. Since there are more men than women in our dataset, the ATE estimate is a little bit closer to the man’s ATE. This is called a non-parametric estimate, since it places no assumption on how the data was generated.

If we control for sex using regression, we will add the assumption of linearity. Regression will also partition the data into man and woman and estimate the effect on both of these groups

It should be

This estimate is done by 1) partitioning the data into confounder cells, in this case, men and women, 2) estimating the effect on each cell and 3) combining the estimate with a weighted average, where the weight is the sample size of the cell or covariate group. If we had exactly the same number of men and women in the data, the ATE estimate would be right in the middle of the ATE of the 2 groups, -2.5. Since there are more men than women in our dataset, the ATE estimate is a little bit closer to the man’s ATE. This is called a non-parametric estimate, since it places no assumption on how the data was generated.

and

If we control for sex using regression, we will add the assumption of linearity. Regression will also partition the data into men and women and estimate the effect on both of these groups

changes highlighted in bold text