How well does the project function work one year in advance?
library(knitr)
library(data.table)
if (requireNamespace("taxstats", quietly = TRUE)){
library(taxstats)
sample_files_all <- get_sample_files_all()
} else {
install.packages("taxstats", repos = "https://hughparsonage.github.io/drat/", type = "source")
library(taxstats)
sample_files_all <- get_sample_files_all()
}
library(grattan)
library(dtplyr)
library(dplyr)
library(magrittr)
library(ggplot2)
library(scales)
library(broom)To test the performance of our projection, we consider the projection of the 2012-13 sample file by one year with the actual 2013-14 sample file.
sample_file_1314_projected <-
sample_file_1213 %>%
copy %>%
mutate(WEIGHT = 50) %>%
project(h = 1L,
fy.year.of.sample.file = "2012-13",
.recalculate.inflators = TRUE) %>%
.[]data.table(the_source = c("Actual",
"Projected"),
n_persons = c(nrow(sample_file_1314) * 50,
sum(sample_file_1314_projected$WEIGHT)),
avg_Taxable_Income = c(mean(sample_file_1314$Taxable_Income),
mean(sample_file_1314_projected$Taxable_Income)),
avg_Sw = c(mean(sample_file_1314$Sw_amt),
mean(sample_file_1314_projected$Sw_amt))
) %>%
melt.data.table(id.vars = "the_source") %>%
group_by(variable) %>%
mutate(value_rel = value / first(value)) %>%
# dcast.data.table(variable ~ the_source) %>%
ggplot(aes(x = variable, y = value_rel, fill = the_source)) +
geom_bar(stat = "identity", position = "dodge") +
geom_text(aes(label = comma(round(value))),
position = position_dodge(width = 0.9),
hjust = 1.02) +
coord_flip() +
theme(legend.position = "top")
conf_int_of_t.test <- function(variable){
t_test <- t.test(sample_file_1314[[variable]],
sample_file_1314_projected[[variable]])
t_test %>%
tidy(.) %>%
mutate(var = variable) %>%
as.data.table
}
c("Sw_amt",
"Net_rent_amt",
"Net_CG_amt",
"Tot_inc_amt",
"Tot_ded_amt",
"Taxable_Income") %>%
lapply(conf_int_of_t.test) %>%
rbindlist %>%
.[, list(var, conf.low, conf.high, p.value)] %>%
ggplot(aes(x = var,
ymin = conf.low,
ymax = conf.high,
color = p.value > 0.05)) +
geom_errorbar() +
geom_hline(yintercept = 0)