Routines for Performing Empirical Calibration of Observational Study Estimates
Calibrate confidence intervals
Calibrate the log likelihood ratio
Calibrate the p-value
Compare EASE of correlated sets of estimates
Compute the (traditional) confidence interval
Compute the (traditional) p-value
Fit the null distribution using non-normal log-likelihood approximatio...
Simulate (negative) controls
Simulate survival data for MaxSPRT computation
Plot the MCMC trace
Compute critical values for Binomial data
Compute critical values for Poisson data
Compute critical values for Poisson regression data
Compute the expected absolute systematic error
Convert empirical null distribution to systematic error model
EmpiricalCalibration: Routines for Performing Empirical Calibration of...
Evaluate confidence interval calibration
Fit the null distribution using MCMC
Fit the null distribution
Fit a systematic error model
Create a calibration plot
Plot the effect of the calibration
Create a confidence interval calibration plot
Plot true and observed values
Plot the effect of the CI calibration
Create a confidence interval coverage plot
Plot the systematic error model
Plot the expected type 1 error as a function of standard error
Create a forest plot
Routines for performing empirical calibration of observational study estimates. By using a set of negative control hypotheses we can estimate the empirical null distribution of a particular observational study setup. This empirical null distribution can be used to compute a calibrated p-value, which reflects the probability of observing an estimated effect size when the null hypothesis is true taking both random and systematic error into account. A similar approach can be used to calibrate confidence intervals, using both negative and positive controls. For more details, see Schuemie et al. (2013) <doi:10.1002/sim.5925> and Schuemie et al. (2018) <doi:10.1073/pnas.1708282114>.
Useful links