Skip to content

Conversation

@poornimaramesh
Copy link
Collaborator

No description provided.

@dougj892 dougj892 marked this pull request as ready for review March 12, 2025 15:19
Copy link
Collaborator

@dougj892 dougj892 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hey - Just had one small comment

return gdf["neighbor_count"].values


def randomly_sample_rooftops(rooftop_neighbour_count: list | np.ndarray,
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hi @poornimaramesh - It looks like this function samples rooftops with probability proportional to the number of neighbors. In our testing in India, we actually found that using simple random sampling within PSUs slightly outperforms other more complicated forms of sampling in terms of ensuring equal probability of household selection and reduction of bias. The difference in performance was very minor, but considering SRS is a bit easier for folks to understand and implement I would prefer for it to be the default. Would it be possible to either a) change the name of this function to make clear that this is implementing sampling using prob proportional to number of neighbors or b) change to SRS?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants