You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Clarification about counter value in generate_pseudo_pop() #138
Hello! Falco and I are working on the school shooting project and we are trying to understand the value of generate_pseudo_pop(), specifically the item counter. Is counter, a vector of length equal to the length of the pseudopopulation, representing how many times each real observation (that is untrimmed by the matching algorithm) is matched in the pseudopopulation?
We're trying to clarify this because 77 out of 48,949 values in counter are ≥1000, with the three largest values being 17,649, 18,645, and 23,473.
Also, if I run sum(counter), should I get length(counter) * {number of GPS bins}? And is {number of GPS bins} = (max(gps) - min(gps)) / delta_n? Unfortunately, I'm not seeing that this is true in my code.
Notation: gps, like counter, is one of the items in the value of generate_pseudo_pop(). delta_n is the parameter I input into generate_pseudo_pop().
I'm trying to understand the GPS matching algorithm (described on page 10 of Wu, Xiao, et al. "Matching on generalized propensity scores with continuous exposures." arXiv preprint arXiv:1812.06575 (2018)).
Activity
m-qin commentedon Jun 30, 2022
Also, if I run
sum(counter), should I getlength(counter)* {number of GPS bins}? And is {number of GPS bins} = (max(gps)-min(gps)) /delta_n? Unfortunately, I'm not seeing that this is true in my code.Notation:
gps, likecounter, is one of the items in the value of generate_pseudo_pop().delta_nis the parameter I input into generate_pseudo_pop().I'm trying to understand the GPS matching algorithm (described on page 10 of Wu, Xiao, et al. "Matching on generalized propensity scores with continuous exposures." arXiv preprint arXiv:1812.06575 (2018)).