- Avoid an error during testing, on R built
- Fix an issue introduced in v0.2.12 that led to an unexpected error in
shape() when 1) at least two
group_names are specified in an order other than alphabetic and 2) geographic
modifier_data is used.
- Allow modeling of unobserved groups using aggregated data. The previous behavior was to drop rows in
aggregate_data indicating zero trials. (They don’t represent item responses.) Preserving them has the effect that unobserved groups, defined partially or entirely by the values of the grouping variables in zero-trial rows in
aggregate_data, can be included in a model.
- Fix an unexpected error when 1)
aggregate_data is used without
item_data, 2) no demographic groups are specified via
group_names, and 3) geographic
modifier_data is used.
- Fix the check for missing
modifier_data must cover all combinations of the geo and time variables in the item response data (individual or aggregated), but because of a bug in the validation of the geographic data, this requirement was not always enforced. In some cases a warning would appear instead of an error.
- Add poststratification over posterior samples (closes #21).
shape() now accepts aggregated item response data unaccompanied by individual-level item response data. The
item_names arguments are no longer required.
- Add a
max_raked_weight argument to
shape() for trimming raked weights. Note that trimming occurs before raked weights are rescaled to have mean 1, and the rescaled weights can be larger than
- Remove the unused function
- Remove Rcpp dependency by rewriting
dichotomize() in R.
- Avoid estimating models (using RStan) during tests, with the goal of rendering moot variation in build environments. This addresses a test failure during CRAN’s r-release-osx-x86_64 build.
- Switch from compiling Stan models at install time to compiling them at runtime, avoiding an Rcpp module issue.
model argument to
dgmrp() taking for reuse a previously compiled Stan model, as found in the
@stanmodel slot of a
version argument to
dgmrp() can be used to specify arbitrary
.stan files on the disk in addition to those included with the package.
get_item_n() methods properly accepts a vector of variable names when combined with
- Improve Stan models for shorter run times
dgmrp() for fitting single-issue MRP models with hierarchical covariates
- Add class
dgmrp_fit for models fitted with
dgmrp(), inheriting from a new virtual class
dgirt() now returns a
dgirt_fit-class object that also inherits from
- Package renamed dgo: Dynamic Estimation of Group-level Opinion
- Tweaks to pass CRAN checks: clean up examples and docs
- Use roxygen2 for classes, methods, and
- Fix checks on
S related to
group_names change in 0.2.5
- Fix Rcpp module issue from 0.2.6 (
Error in .doLoadActions(where, attach))
- Fix error in
- Fix path in
group_names is no longer required. If omitted, the geographic variable given by
geo_name will define groups.
aggregate_item_names is no longer required. It defaults to the observed values of the
item column in
raking argument to
strata_names. It takes a formula or list of formulas and allows more complicated preweighting.
id_vars argument to
shape() specifies variables to be kept in
aggregate_data may include geographic areas, demographics, or time periods that don’t appear in
- Fix: use a smaller epsilon than the default in survey::rake() for convergence with non-frequency weights.
plot_rhats() for model checking.
get_time_elapsed gives model run times. These also appear in