Libraries
library(knitr)
library(forecast)
## Warning: package 'forecast' was built under R version 3.6.2
## Registered S3 method overwritten by 'xts':
## method from
## as.zoo.xts zoo
## Registered S3 method overwritten by 'quantmod':
## method from
## as.zoo.data.frame zoo
library(fpp2)
## Warning: package 'fpp2' was built under R version 3.6.2
## ── Attaching packages ─────────────────────────────────────────────────────────────── fpp2 2.4 ──
## ✓ ggplot2 3.3.2 ✓ expsmooth 2.3
## ✓ fma 2.4
## Warning: package 'ggplot2' was built under R version 3.6.2
##
library(dplyr)
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
Prices of Bushels of Corn, US
corn = read.csv("/Users/nelsonwhite/Documents/ms applied economics/Predictive Analytics:Forecasting/week 7 discussion/corn-prices-historical-chart-data.csv")
corn.ts = ts(corn[,2], start = c(01,07,1959), frequency = 52)
autoplot(corn.ts, ylab = "Bushel of Corn, Dollars", main = "Prices of Corn over Time")
Analysis
corn.n1=nnetar(corn.ts,lambda="auto",size=1)
corn.n2=nnetar(corn.ts,lambda="auto",size=2)
corn.n3=nnetar(corn.ts,lambda="auto",size=3)
corn.n1.fcast = forecast(corn.n1)
corn.n2.fcast = forecast(corn.n2)
corn.n3.fcast = forecast(corn.n3)
autoplot(corn.n1.fcast)
accuracy(corn.n1.fcast)
## ME RMSE MAE MPE MAPE
## Training set 0.004282108 0.05958127 0.03327117 0.02966489 1.045119
## MASE ACF1
## Training set 0.1330543 0.3358256
autoplot(corn.n2.fcast)
accuracy(corn.n2.fcast)
## ME RMSE MAE MPE MAPE
## Training set 0.001745044 0.05124571 0.03013714 0.002366774 0.9772269
## MASE ACF1
## Training set 0.120521 0.09862
autoplot(corn.n3.fcast)
accuracy(corn.n3.fcast)
## ME RMSE MAE MPE MAPE
## Training set 0.0009218864 0.04975322 0.02960908 -0.0006694709 0.9693828
## MASE ACF1
## Training set 0.1184093 0.04164264