2015-16 size of dividends credit

library(data.table)
library(SampleFile1516)
library(SampleFile1415)
library(grattan)
library(hutils)
library(magrittr)
library(ggplot2)
library(grattanCharts)

## 
## Attaching package: 'grattanCharts'

## The following object is masked from 'package:datasets':
## 
##     Orange

library(scales)

library(knitr)
opts_chunk$set(results = "hide")

s1516 <- as.data.table(sample_file_1516)
s1516[, Sex := c("Male", "Female")[Gender + 1L]]

s1415 <- as.data.table(sample_file_1415)
s1415[, Sex := c("Male", "Female")[Gender + 1L]]

s1516[, tax := income_tax(Taxable_Income, "2015-16", .dots.ATO = copy(s1516))]
s1516[, tax_no_frnk := tax + (tax <= Dividends_franking_cr_amt) * Dividends_franking_cr_amt]
s1516[, isAffected := tax_no_frnk > tax]
s1516[, delta := tax_no_frnk - tax]

s1415[, tax := income_tax(Taxable_Income, "2014-15", .dots.ATO = copy(s1415))]
s1415[, tax_no_frnk := tax + (tax <= Dividends_franking_cr_amt) * Dividends_franking_cr_amt]
s1415[, isAffected := tax_no_frnk > tax]
s1415[, delta := tax_no_frnk - tax]

opts_chunk$set(results = "markup")

# Percent women:
s1415[, .(isAffected = mean(isAffected)), keyby = "Sex"]

##       Sex isAffected
## 1: Female 0.08014375
## 2:   Male 0.05970204

s1516[, .(isAffected = mean(isAffected)), keyby = "Sex"]

##       Sex isAffected
## 1: Female 0.08082132
## 2:   Male 0.05997154

# Affected and women
s1516[(isAffected), mean(Gender)]

## [1] 0.5582023

# 0.5582023

# Affected and over 60
s1415[(isAffected & Gender == 1), mean(age_range <= 2)]

## [1] 0.6792155

# 0.6792155

# Average sizes by gender
s1516[, .(avgAffected = mean(Dividends_franking_cr_amt)), keyby = "Gender"]

##    Gender avgAffected
## 1:      0    718.6282
## 2:      1    632.0685

s1516[Dividends_franking_cr_amt > 0, .(avgAffected = mean(Dividends_franking_cr_amt)), keyby = "Gender"]

##    Gender avgAffected
## 1:      0    3447.503
## 2:      1    2888.947

# Average size of dividends credit among those affected
s1516[(isAffected), .(avgAffected = mean(Dividends_franking_cr_amt)), keyby = "Gender"]

##    Gender avgAffected
## 1:      0    2636.391
## 2:      1    2306.922

# Average change in tax if Div_cr non-refundable
s1516[, .(avgDelta = mean(delta)), keyby = "Sex"]

##       Sex avgDelta
## 1: Female 186.4485
## 2:   Male 158.1084

# Average change in tax if Div_cr non-refundable -- among those affected
s1516[(isAffected), .(avgDelta = mean(delta)), keyby = "Sex"]

##       Sex avgDelta
## 1: Female 2306.922
## 2:   Male 2636.391