In FPRgroup/FPEMcountry: fpemlocal

knitr::opts_chunk$set(echo = TRUE)
knitr::opts_knit$set(root.dir = rprojroot::find_rstudio_root_file())
# knitr::opts_knit$get("root.dir")  # alternative to the previous line
# the default autosave location will depend on this being setup
options(warn=-1)

library(fpemlocal)

Introduction

In this vignette we obtain estimates for married women with custom user data.

1. Fit a one country model fit_fp_c
2. Calculate point estimates for indicators calc_fp_c
3. Plot the point estimates against the survey data plot_fp_c

1. Fit a one country model

Fit the one-country family planning estimation model with the function fit_fp_c. Supply the file path of the .csv file containing your country survey data to the argument surveydata_filepath. Next supply the UNPD country code known as the division_numeric_code. See ??divisions for UNPD country codes and other divisions. Specify the model of your choice with the argument is_in_union. There are two versions of this model, one for in-union women and another for not-in-union women denoted "Y" and "N" respectively. Lastly, specify the years of estimates to be returned. Note: The function will not filter the survey data based on these years. All years of available survey data will be used.

fit <- fit_fp_c(
  surveydata_filepath = "data-raw/manuscript_example_data/afghanistan_4_married_example.csv",
  division_numeric_code = 4,
  is_in_union = "Y",
  first_year = 1970,
  last_year = 2030
)

2. Calculate point estimates for indicators

Calculate point estimates for family planning indicators with the function calc_fp_c. Supply the fit object from fit_fp_c to the argument fit. Read in your population count dataset. Then supply the dataset to the argument population_data.

population_data <- read.csv("data-raw/manuscript_example_data/afghanistan_4_married_popdata_example.csv")
results <- calc_fp_c(fit = fit,
                     population_data = population_data)

A set of results here consist of the following family planning indicators

results$Y %>% names

The point estimates for each indicator are long-format tibbles. Let's take a look at the tibble for the indicator contraceptive_use_modern

results$Y$contraceptive_use_modern

3. Plot estimates and survey data

fpemlocal also includes a function named plot_fp_c to plot the calculated point estimates against the survey data. The arguments to this function are, the fit object from step 1, the results from step 2, and a vector of indicator names. The vector of indicator names corresponds to the names which appear in the results from step 2.

plot_fp_c(
  fit,
  results,
  indicators = c(
    "unmet_need_any",
    "contraceptive_use_modern",
    "contraceptive_use_traditional",
    "contraceptive_use_any"
    )
  )

FPRgroup/FPEMcountry documentation built on April 24, 2023, 4:32 p.m.