WDFC Linear Forecast

Libraries

library(fpp2)

## Warning: package 'fpp2' 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

## ── Attaching packages ──────────────────────────────────────────────────────────────────────────── fpp2 2.4 ──

## ✓ ggplot2   3.3.2     ✓ fma       2.4  
## ✓ forecast  8.13      ✓ expsmooth 2.3

## Warning: package 'ggplot2' was built under R version 3.6.2

## Warning: package 'forecast' 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

library(knitr)

Download and trim data

w=read.csv("/Users/nelsonwhite/Documents/ms applied economics/Predictive Analytics:Forecasting/week 1 discussion/WDFC.csv")

w=ts(w[,2],start=c(2018,10,23),end=c(2020,10,22),frequency=250)

WDFC over time

autoplot(w)

Linear model prediction

model1 = tslm(w~trend)

 autoplot(w, series="Data") +
     autolayer(fitted(model1), series="Fit") +
     autolayer(forecast(model1), series="Projection") +
     xlab("Year") + ylab("Price") +
     ggtitle("WDFC Linear Forecast")

Check residials and accuracy

checkresiduals(model1)

## 
##  Breusch-Godfrey test for serial correlation of order up to 100
## 
## data:  Residuals from Linear regression model
## LM test = 423.72, df = 100, p-value < 2.2e-16

accuracy(model1)

##                         ME     RMSE      MAE        MPE    MAPE     MASE
## Training set -3.433696e-16 10.08061 7.353367 -0.2821188 3.91959 1.228065
##                   ACF1
## Training set 0.8884767