Policy Frontier Plot

I found a great tutorial here: http://tutorials.iq.harvard.edu/R/Rgraphics/Rgraphics.html

Step 1: Load the data. These data were generated by a numerical simulaton performed in Python.

require(ggplot2)

## Loading required package: ggplot2

## Warning: package 'ggplot2' was built under R version 3.3.3

setwd("D:/Dropbox/Research/PriceDispersionOMP/Simulation")

NGDP<- data.frame(read.csv("ngdp1_Frontier.csv")[c("sig_r", "sig_y", "sig_pi")])
NGDP$rule<-"NGDP Rule"

IT <- data.frame(read.csv("Inf_Rule_2_Frontier.csv")[c("sig_r", "sig_y", "sig_pi")])
IT$rule <- "Inflation Rule"

TRule_unemp = data.frame(read.csv("u2_Frontier.csv")[c("sig_r", "sig_y", "sig_pi")])
TRule_unemp$rule <-"Taylor Rule Unemp"

mydata <- rbind(NGDP, IT ,TRule_unemp)

#Get rid of outliers
mydata<-subset(mydata, sig_y<4 & sig_pi<6)

Step 2: Create a basic plot that shows the plolicy frontier of 3 different monetary policy rulies

p_basic <- ggplot(mydata, aes(x=sig_pi, y=sig_y)) +
  geom_line(aes(linetype=rule))+
  scale_x_continuous(name="Standard deviation of inflation")+
  scale_y_continuous(name="Standard deviation of GDP growth")
p_basic

Step 3: We have no idea how volatile the policy interest rate is in these 3 scenarios. It would be nice to see this on the graph.

p1 <- ggplot(mydata, aes(x=sig_pi, y=sig_y)) +
  geom_point(aes(color=sig_r, shape=rule)) +
  scale_x_continuous(name="Standard deviation of inflation") +
  scale_y_continuous(name="Standard deviation of GDP growth") +
  scale_color_continuous(name="st. dev of policy rate",
                         low = "blue", high="red")
p1