EmpiricalCalibration3.1.4 package

Routines for Performing Empirical Calibration of Observational Study Estimates

calibrateConfidenceInterval

Calibrate confidence intervals

calibrateLlr

Calibrate the log likelihood ratio

calibrateP

Calibrate the p-value

compareEase

Compare EASE of correlated sets of estimates

computeTraditionalCi

Compute the (traditional) confidence interval

computeTraditionalP

Compute the (traditional) p-value

fitNullNonNormalLl

Fit the null distribution using non-normal log-likelihood approximatio...

simulateControls

Simulate (negative) controls

simulateMaxSprtData

Simulate survival data for MaxSPRT computation

plotMcmcTrace

Plot the MCMC trace

computeCvBinomial

Compute critical values for Binomial data

computeCvPoisson

Compute critical values for Poisson data

computeCvPoissonRegression

Compute critical values for Poisson regression data

computeExpectedAbsoluteSystematicError

Compute the expected absolute systematic error

convertNullToErrorModel

Convert empirical null distribution to systematic error model

EmpiricalCalibration-package

EmpiricalCalibration: Routines for Performing Empirical Calibration of...

evaluateCiCalibration

Evaluate confidence interval calibration

fitMcmcNull

Fit the null distribution using MCMC

fitNull

Fit the null distribution

fitSystematicErrorModel

Fit a systematic error model

plotCalibration

Create a calibration plot

plotCalibrationEffect

Plot the effect of the calibration

plotCiCalibration

Create a confidence interval calibration plot

plotTrueAndObserved

Plot true and observed values

plotCiCalibrationEffect

Plot the effect of the CI calibration

plotCiCoverage

Create a confidence interval coverage plot

plotErrorModel

Plot the systematic error model

plotExpectedType1Error

Plot the expected type 1 error as a function of standard error

plotForest

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>.

  • Maintainer: Martijn Schuemie
  • License: Apache License 2.0
  • Last published: 2025-02-14