Is there a preferred method for estimating the required sample size for testing the accuracy of a gold test against a new under trial test.
My denominator includes only the positives from the gold test. So the question is how many of the gold tests should be done to determine the accuracy of the the trial test. I am aiming for at least 85-90% or higher accuracy (sensitivity).
Basically if sensitivity is my only concern then what should be my approach to sample size estimation?
Related Posts with sample size for test accuracy
Inconsistency between rdrobust and rdplot resultsHi everyone, I use the rdrobust/rdplot package in stata to estimate a simple RD model with time as t…
Regarding generating year-sum for each variableDear, I have a question. I have a panel data looks like below. Code: country year x y "I…
IV estimates with low mean values of the outcome?Dear all, I am examining the effects of maternal education on low bitrth weights using a fuzzy regr…
Removing duplicate ID's by creating additional variableHi All, I would like to change this dataset such that each ID only appears one time, with additiona…
Combine the panel data with External Covariates.I have DHS data which have hhid(household id), BMI(Body Mass Index), and Year(2005, 2010, 2015. The …
Subscribe to:
Post Comments (Atom)
0 Response to sample size for test accuracy
Post a Comment