MRPWorkflow objects

A MRPWorkflow object is an R6 object created by the mrp_workflow() function. This class provides methods for all steps in the workflow, from data preparation and visualization to model fitting.

Format

An R6 generator object.

Methods

Public methods

• MRPWorkflow$new()

• MRPWorkflow$metadata()

• MRPWorkflow$data()

• MRPWorkflow$preprocess()

• MRPWorkflow$link_acs()

• MRPWorkflow$load_pstrat()

• MRPWorkflow$demo_bars()

• MRPWorkflow$covar_hist()

• MRPWorkflow$sample_size_map()

• MRPWorkflow$outcome_plot()

• MRPWorkflow$outcome_map()

• MRPWorkflow$estimate_plot()

• MRPWorkflow$estimate_map()

• MRPWorkflow$create_model()

• MRPWorkflow$pp_check()

• MRPWorkflow$compare_models()

• MRPWorkflow$clone()

---

Method new()

Initializes the MRPWorkflow object, setting up necessary fields for data processing and model fitting.

Usage

MRPWorkflow$new()

Returns

A new MRPWorkflow object.

---

Method metadata()

Retrieves the metadata associated with the current workflow, including information about time variables, family, and special cases.

Usage

MRPWorkflow$metadata()

---

Method data()

Retrieves the preprocessed sample data that has been prepared for MRP analysis.

Usage

MRPWorkflow$data()

---

Method preprocess()

Preprocesses the input sample data by cleaning, validating, and preparing it for MRP analysis. This includes handling time-varying data, aggregated data, and special cases.

Usage

MRPWorkflow$preprocess(
  data,
  is_timevar = FALSE,
  is_aggregated = FALSE,
  special_case = NULL,
  family = NULL,
  zip_threshold = 0,
  state_threshold = 0
)

Arguments

data

Input sample data to be preprocessed

is_timevar

Logical indicating whether the data contains time-varying components

is_aggregated

Logical indicating whether the data is already aggregated

special_case

Character string specifying special case handling (e.g., "covid", "poll")

family

Character string specifying the model family (e.g., "binomial", "normal")

zip_threshold

Numeric value specifying the minimum number of records required for a ZIP code to be included in the analysis (default is 0)

state_threshold

Numeric value specifying the minimum number of records required for a state to be included in the analysis (default is 0)

---

Method link_acs()

Links the preprocessed sample data to ACS poststratification data based on geographic and demographic variables.

Usage

MRPWorkflow$link_acs(link_geo = NULL, acs_year = 2023)

Arguments

link_geo

Character string specifying the geographic level for linking (e.g., "state", "county", "zip")

acs_year

Numeric value specifying the ACS year to use for poststratification data

---

Method load_pstrat()

Loads and processes custom poststratification data instead of using ACS data. This method validates the data and prepares it for MRP analysis.

Usage

MRPWorkflow$load_pstrat(pstrat_data, is_aggregated = FALSE)

Arguments

pstrat_data

Custom poststratification data to be loaded

is_aggregated

Logical indicating whether the poststratification data is already aggregated

---

Method demo_bars()

Creates bar plots comparing demographic distributions between input survey data and target population data.

Usage

MRPWorkflow$demo_bars(demo, file = NULL, ...)

Arguments

demo

Character string specifying the demographic variable to plot

file

Optional file path to save the plot

...

Additional arguments passed to ggsave

Returns

A ggplot object or patchwork object showing demographic comparisons

---

Method covar_hist()

Creates histogram plots showing the distribution of geographic covariates across zip codes. Only available for COVID data.

Usage

MRPWorkflow$covar_hist(covar, file = NULL, ...)

Arguments

covar

Character string specifying the covariate. Options: "edu", "poverty", "employ", "income", "urban", "adi"

file

Optional file path to save the plot

...

Additional arguments passed to ggsave

Returns

A ggplot object showing the covariate distribution histogram

---

Method sample_size_map()

Creates interactive choropleth maps showing data distribution with respect to geography.

Usage

MRPWorkflow$sample_size_map(file = NULL)

Arguments

file

Optional file path to save the plot

Returns

A highcharter map object showing sample size distribution

---

Method outcome_plot()

Creates plots showing the distribution of outcome measures over time (for time-varying data) or as static distributions (for cross-sectional data).

Usage

MRPWorkflow$outcome_plot(file = NULL, ...)

Arguments

file

Optional file path to save the plot

...

Additional arguments passed to ggsave

Returns

A ggplot object showing the outcome measure distribution

---

Method outcome_map()

Creates maps showing average outcome measure by geography for cross-sectional data, or highest/lowest weekly average for time-varying data.

Usage

MRPWorkflow$outcome_map(summary_type = NULL, file = NULL)

Arguments

summary_type

Character string for time-varying data: "max" or "min"

file

Optional file path to save the map

Returns

A highcharter map object showing outcome measures by geography

---

Method estimate_plot()

Creates plots showing MRP estimates for different subgroups, either over time (for time-varying data) or as static estimates (for cross-sectional data).

Usage

MRPWorkflow$estimate_plot(
  model,
  group = NULL,
  interval = 0.95,
  show_caption = TRUE,
  file = NULL,
  ...
)

Arguments

model

Fitted MRPModel object

group

Character string specifying the demographic group for plotting

interval

Confidence interval or standard deviation for the estimates (default is 0.95)

show_caption

Logical indicating whether to show the caption in the plot (default is TRUE)

file

Optional file path to save the plot

...

Additional arguments passed to ggsave

Returns

A ggplot object showing the group estimates

---

Method estimate_map()

Creates interactive choropleth maps showing MRP estimates by geographic regions.

Usage

MRPWorkflow$estimate_map(
  model,
  geo = NULL,
  time_index = NULL,
  interval = 0.95,
  file = NULL,
  ...
)

Arguments

model

Fitted MRPModel object

geo

Character string specifying the geographic level for mapping

time_index

Numeric value specifying the time index for time-varying data

interval

Confidence interval or standard deviation for the estimates (default is 0.95)

file

Optional file path to save the map

...

Additional arguments

Returns

A highcharter map object showing MRP estimates by geography

---

Method create_model()

Creates a new MRPModel object with validated effects specification and prepared data for Bayesian model fitting.

Usage

MRPWorkflow$create_model(model_spec)

Arguments

model_spec

List containing model effects specification including intercept, fixed effects, varying effects, and interactions

Returns

A new MRPModel object

---

Method pp_check()

Creates posterior predictive check plots to assess model fit by comparing observed data to replicated data from the posterior predictive distribution.

Usage

MRPWorkflow$pp_check(model, file = NULL, ...)

Arguments

model

Fitted MRPModel object

file

Optional file path to save the plot

...

Additional arguments passed to ggsave

---

Method compare_models()

Compares multiple fitted MRP models using leave-one-out cross-validation to assess relative model performance.

Usage

MRPWorkflow$compare_models(..., suppress = NULL)

Arguments

...

Multiple MRPModel objects to compare

suppress

Character string specifying output to suppress during comparison

Returns

A data frame summarizing the comparison results

---

Method clone()

The objects of this class are cloneable with this method.

Usage

MRPWorkflow$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.