Loading the packages

library(forecast)

Reading the data

wines <- read.csv(file="AustralianWines.csv")

Excluding the missing data.

wines <- na.omit(wines)

Data partitioning

fortified <- ts(wines$Fortified, start=c(1980,1), end=c(1994,12), frequency = 12)

train.fortified <- fortified[1:178]
train.fortified <- ts(train.fortified, start=c(1980,1), frequency=12 )

test.fortified <- fortified[179:180]
test.fortified <- ts(test.fortified, start=c(1994,11), frequency=12 )


red <- ts(wines$Red, start=c(1980,1), end=c(1994,12), frequency = 12)

train.red <- red[1:178]
train.red <- ts(train.red, start=c(1980,1), frequency=12 )

test.red <- red[179:180]
test.red <- ts(test.red, start=c(1994,11), frequency=12 )


rose <- ts(wines$Rose, start=c(1980,1), end=c(1994,12), frequency = 12)

train.rose <- rose[1:178]
train.rose <- ts(train.rose, start=c(1980,1), frequency=12 )

test.rose <- rose[179:180]
test.rose <- ts(test.rose, start=c(1994,11), frequency=12 )


sparkling <- ts(wines$sparkling, start=c(1980,1), end=c(1994,12), frequency = 12)

train.sparkling <- sparkling[1:178]
train.sparkling <- ts(train.sparkling, start=c(1980,1), frequency=12 )

test.sparkling <- sparkling[179:180]
test.sparkling <- ts(test.sparkling, start=c(1994,11), frequency=12 )


sweet.white <- ts(wines$Sweet.white, start=c(1980,1), end=c(1994,12), frequency = 12)

train.sweet.white <- sweet.white[1:178]
train.sweet.white <- ts(train.sweet.white, start=c(1980,1), frequency=12 )

test.sweet.white <- sweet.white[179:180]
test.sweet.white <- ts(test.sweet.white, start=c(1994,11), frequency=12 )


dry.white <- ts(wines$Dry.white, start=c(1980,1), end=c(1994,12), frequency = 12)

train.dry.white <- dry.white[1:178]
train.dry.white <- ts(train.dry.white, start=c(1980,1), frequency=12 )

test.dry.white <- dry.white[179:180]
test.dry.white <- ts(test.dry.white, start=c(1994,11), frequency=12 )

Naive method

The R package forecast has the function naive which can be used to generate forecasting using naive method. We are going to generate naive forecasts for January and February 1995, for each of the six wine type with the following code:

Fortified

naive.fortified <- naive(fortified, h=2)
naive.fortified$mean
##       Jan  Feb
## 1995 2467 2467

Red

naive.red <- naive(red, h=2)
naive.red$mean
##       Jan  Feb
## 1995 2684 2684

Rose

naive.rose <- naive(rose, h=2)
naive.rose$mean 
##      Jan Feb
## 1995  83  83

Sparkling

naive.sparkling <- naive(sparkling, h=2)
naive.sparkling$mean 
##       Jan  Feb
## 1995 5999 5999

Sweet white

naive.sweet.white <- naive(sweet.white, h=2)
naive.sweet.white$mean
##      Jan Feb
## 1995 394 394

Dry white

naive.dry.white <- naive(dry.white, h=2)
naive.dry.white$mean
##       Jan  Feb
## 1995 5725 5725

Seasonal naive method

We also perform the forecast using the seasonal naive method, as follows:

Fortified

naive.fortified <- snaive(fortified, h=2)
naive.fortified$mean
##       Jan  Feb
## 1995 1154 1568

Red

naive.red <- snaive(red, h=2)
naive.red$mean
##       Jan  Feb
## 1995 1041 1728

Rose

naive.rose <- snaive(rose, h=2)
naive.rose$mean 
##      Jan Feb
## 1995  44  47

Sparkling

naive.sparkling <- snaive(sparkling, h=2)
naive.sparkling$mean 
##       Jan  Feb
## 1995 1197 1968

Sweet white

naive.sweet.white <- snaive(sweet.white, h=2)
naive.sweet.white$mean
##      Jan Feb
## 1995 150 280

Dry white

naive.dry.white <- snaive(dry.white, h=2)
naive.dry.white$mean
##       Jan  Feb
## 1995 2265 3685