ETS Models

library(forecast)

## Registered S3 method overwritten by 'quantmod':
##   method            from
##   as.zoo.data.frame zoo

library(fma)
library(fpp2)

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

## ✓ ggplot2   3.3.5     ✓ expsmooth 2.3

## 

library(readxl)
crudeoil <- read_excel("~/Desktop/BC 22Spr/crudeoil.xlsx", 
    sheet = "Sheet1 (3)")
View(crudeoil)
str(crudeoil)

## tibble [145 × 2] (S3: tbl_df/tbl/data.frame)
##  $ time          : chr [1:145] "2009 01" "2009 02" "2009 03" "2009 04" ...
##  $ crudeOilImport: num [1:145] 317275 262339 303897 285934 281147 ...

#convert it to a time serie and plot
myts=ts(crudeoil$crudeOilImport,frequency=12,start=c(2009,1))
plot(myts)

#ANN
mye=ets(myts,model="ANN")
plot(mye)

mye

## ETS(A,N,N) 
## 
## Call:
##  ets(y = myts, model = "ANN") 
## 
##   Smoothing parameters:
##     alpha = 0.3651 
## 
##   Initial states:
##     l = 294004.805 
## 
##   sigma:  15407.75
## 
##      AIC     AICc      BIC 
## 3521.974 3522.144 3530.904

fe=forecast(mye)
accuracy(mye)

##                     ME     RMSE      MAE       MPE     MAPE      MASE
## Training set -2231.259 15301.12 11431.29 -1.318732 4.877614 0.6395479
##                    ACF1
## Training set -0.1031853

plot(fe,main="ANN")

#AAN
mye1=ets(myts,model="AAN")
plot(mye1)

mye1

## ETS(A,A,N) 
## 
## Call:
##  ets(y = myts, model = "AAN") 
## 
##   Smoothing parameters:
##     alpha = 0.3352 
##     beta  = 1e-04 
## 
##   Initial states:
##     l = 297256.9026 
##     b = -446.2055 
## 
##   sigma:  15375.29
## 
##      AIC     AICc      BIC 
## 3523.320 3523.752 3538.204

fe=forecast(mye1)
accuracy(mye1)

##                     ME     RMSE      MAE        MPE     MAPE      MASE
## Training set -1167.603 15161.74 11379.31 -0.8698161 4.834512 0.6366398
##                    ACF1
## Training set -0.0686389

plot(fe,main="AAN")

#AAA
mye2=ets(myts,model="AAA")
plot(mye2)

mye2

## ETS(A,A,A) 
## 
## Call:
##  ets(y = myts, model = "AAA") 
## 
##   Smoothing parameters:
##     alpha = 0.5274 
##     beta  = 0.0049 
##     gamma = 1e-04 
## 
##   Initial states:
##     l = 290409.3179 
##     b = 40.464 
##     s = 1152.286 -7879.056 -3658.849 -3310.959 10409.33 10768.57
##            3950.87 8748.941 -3127.875 5073.291 -26515.41 4388.867
## 
##   sigma:  10784.21
## 
##      AIC     AICc      BIC 
## 3431.566 3436.385 3482.170

fe=forecast(mye2)
accuracy(mye2)

##                     ME     RMSE      MAE        MPE     MAPE      MASE
## Training set -975.2868 10171.84 7996.505 -0.5523848 3.412448 0.4473815
##                     ACF1
## Training set -0.06320921

plot(fe,main="AAA")