-
Corrected code that produced warning
length > 1 in coercion to logical
-
Update functions that use deprecated functions from
dplyr
package
-
No longer export the following functions:
-
CalcGroupPIPs
,CalcWithinGroupPIPs
, andCalcPIPs
as these should typically be calculated using the functionExtractPIPs
-
ComputePostmeanHnew.approx
andComputePostmeanHnew.exact
as these should typically be calculated using the functionComputePostmeanHnew
-
set_verbose_opts
as this is only called internally
-
-
Expanded function documentation by adding example code
-
allowable values for starting parameter for
r[m]
parameters updated as follows-
no longer truncated to a single value (when
varsel = FALSE
andrmethod = "varying"
) -
can be equal to 0 (when
varsel = TRUE
)
-
-
Error no longer generated if starting values for h.hat are not positive
-
When checking class of an object, use
inherits()
instead ofclass()
-
Added ability to have binomial outcome
family
by implementing probit regression withinkmbayes()
-
Removed computation of the subject-specific effects
h[i]
withinkmbayes()
, as this is not always desired, and greatly slows down model fitting-
This could still be done by setting the option
est.h = TRUE
in thekmbayes
function -
posterior samples of
h[i]
can now be obtained via the post-processingSamplePred
function; alternatively, posterior summaries (mean, variance) can be obtained via the post-processingComputePostmeanHnew
function
-
-
Added ability to use exact estimates of the posterior mean and variance by specifying the argument
method = 'exact'
within the post-processing functions (e.g.,OverallRiskSummaries()
,PredictorResponseUnivar()
)
- Fixed
PredictorResponseBivarLevels()
when argumentboth_pairs = TRUE
(#4)