Method for correction based on testing number

[noamross]

Hello there, great work. I was wondering if you could share your code or methodology for correcting for the number of tests performed in adjusting the number of cases. We are doing some similar work and are trying to compare methods used for this. Thanks! @bbolker

[barretod]

Hi, any updates on this? Looks like you made a major change a week ago and we would like to review that.

[smbartell]

I'm also very curious about that. There's no obvious way to do it. If testing were random, you could just normalize by the number of tests... but testing is not random, it's targeted at those with the highest prior probability of infection based on their symptoms and/or contact with known cases, and as testing expands it inherently dips into people with lower prior probabilities of testing positive. It's difficult to imagine a coherent method of adjustment without detailed information on the testing protocols and case contact patterns in each county.

[barretod]

@k-sys PLEASE push the changes you've made, just noticed you updated the model again two days ago but have decided not to make the model changes public. We're using this model for other countries/regions and this could really help!