- Adding bias-adjusting for exposure misclassification using predictive values.
- Adding start-up message.
- Fix CI for OR in
plot.booted
(@codiewood, #2) - Replaced dot-dot notation by after_stat(density) in
plot.probsens
andplot.booted
misclassification_cov
replaced bymisclassification.cov
- Standard output for
confounders.ext
,confounders.limit
, andconfounders.array
- Suggest
tidyr
instead oftidyverse
- Fix confidence interval values in plotting output of
probsens
. - Fix bug that created integer overflow in rare cases when 2-by-2 table cells were very large.
- Update documentation.
- Provide default values for variables in
mbias
. - Update M-bias plot by using
ggdag
(anddagitty
). - Add beta distribution to probabilistic bias analyses.
- Plotting functions for probabilistic bias analyses.
- Review criteria to remove negative cell counts in probabilistic bias analyses.
- Add citation and DOI.
- Fix formulas in
probsens.conf
.
- Fix selection-bias factor in
selection
which was returning a constant value.
- Fix
bias_parms
inselection
to use a single selection-bias factor (was skipping it and returning NAs).
- Update description of
parms
inconfounders
,confounders.emm
, andconfounders.poly
- New function
confounders.array
, sensitivity analysis for unmeasured confounders based on confounding imbalance among exposed and unexposed (Schneeweiss, 2006) - Deprecated
bias
parameters removed from functionmultidimBias
. Please usebias_parms
instead. - New function
confounders.evalue
, computing E-value to assess bias due to unmeasured confounder (VanderWeele and Ding, 2017) - New function
multiple.bias
allowing to extract 2-by-2 table from anepisensr
object to feed another function (multiple bias analysis)
- Fix bug for distributions and computations of OR/RR in
probsens.conf
- Update distributions in
probsens.irr.conf
- Add example using
probsens.conf
in vignette
- Fix bug when using triangular distribution in
probsens.conf
function for prevalence of exposure among the non-exposed (as producing NaNs) (#1).
- Add
misclassification_cov
for a misclassified covariate (confounder or effect modifier). - As such, this (wrongly) added option into
misclassification
in the previous version is now removed. - Add computation of confidence interval for odds ratio as per Chu et al. for exposure misclassification.
- Use
bias_parms
instead ofbias
inmisclassification
function.
-
Fix bug when building 2-by-2 table.
-
Various formatting improvements in output of
confounders
,confounders.emm
,misclassification
andselection
functions. -
Standardize use of
bias_parms
. -
Add vignette.
-
Selection bias factor now available in output of
selection
function. -
Add bootstrap option
- Fix 2-by-2 tables when variables are provided instead of a matrix.
- Fix R version dependency (R >= 3.2.0)
-
Harmonization of arguments across functions.
-
New distributions added to
probsens
series of functions: constant, logit-logistic, logit-normal, log-logistic, and log-normal. -
Probabilistic analysis of person-time data added with
probsens.irr
for exposure misclassification, andprobsens.irr.conf
for unmeasured confounder. -
Sensitivity analysis to correct for selection bias caused by M bias with
mbias
function, including DAG plot and print function. -
Fix CI formatting.
-
NAMESPACE: add imports to
stats
functions to avoid new R CMD CHECK warnings