TS2 HW1Code.Rmd

---
title: "TS-HW1"
author: "Eric Drew"
date: "2022-10-05"
output: pdf_document
---

```{r setup, include=FALSE}
#LABARR LIBRARIES
library(tseries)
library(tidyverse)
library(forecast)
library(haven)
library(fma)
library(expsmooth)
library(lmtest)
library(zoo)
library(seasonal)
library(ggplot2)
library(seasonalview)
library(aTSA)
library(imputeTS)
library(prophet)
```

```{r data}
data <- read.csv('C:\\Users\\ericd\\OneDrive - North Carolina State University\\Desktop\\AA502\\Time Series II\\Homework 1\\hrl_load_metered.csv')

which(data$mw == 0)
data[5283,'mw'] = (data[5282, 'mw']+data[5284,'mw'])/2
data[14187,'mw'] = (data[14186, 'mw']+data[14187,'mw'])/2

test1 <- read.csv('C:\\Users\\ericd\\OneDrive - North Carolina State University\\Desktop\\AA502\\Time Series II\\Homework 1\\hrl_load_metered - test1.csv')

test2 <- read.csv('C:\\Users\\ericd\\OneDrive - North Carolina State University\\Desktop\\AA502\\Time Series II\\Homework 1\\hrl_load_metered - test2.csv')

data <- rbind(data,test1)


##create time series object
energy <- ts(data$mw, frequency=24)
test <- ts(test2$mw, frequency=24)


training <- subset(energy, end = length(energy)-168)
valid <- subset(energy, start=length(energy)-168)
test <- ts(test2$mw, frequency=24)


```

```{r esm}
HWES.add <- hw(training, seasonal='additive',h=168)
summary(HWES.add)
HWES.add$mean


#BIC: 489730.3 
#MAPE: 3.188799
#MAE: 32.54299
checkresiduals(HWES.add)

#####Additive forecasting
# Calculate prediction errors from forecast
HWES.add.error <- valid - HWES.add$mean

# Calculate prediction error statistics (MAE and MAPE)
HWES.add.MAE <- mean(abs(HWES.add.error))
HWES.add.MAPE <- mean(abs(HWES.add.error)/abs(valid))*100

HWES.add.MAE #126.45
HWES.add.MAPE #15.103

#MULT
HWES.mult <- hw(training, seasonal='multiplicative',h=168)
summary(HWES.mult)
#BIC: 487099.7 
#MAPE: 2.927029
#MAE: 30.49139
checkresiduals(HWES.mult)

#Multiplicative forecasting
#autoplot(forecast::forecast(HWES.add, h = 20)) + autolayer(fitted(HWES.add), series="Fitted") + 
  #ylab("Energy Consumption")

# Calculate prediction errors from forecast
HWES.mult.error <- valid - HWES.mult$mean

# Calculate prediction error statistics (MAE and MAPE)
HWES.mult.MAE <- mean(abs(HWES.mult.error))
HWES.mult.MAPE <- mean(abs(HWES.mult.error)/abs(valid))*100

HWES.mult.MAE #128.397
HWES.mult.MAPE #15.426
```

```{r arima}
training %>% diff(lag = 24) %>% ggtsdisplay()

S.ARIMA <- auto.arima(training, method="ML", seasonal = TRUE)
  summary(S.ARIMA)
#BIC: 294224.8
#MAPE: 3.467278
#MAE: 36.37895
checkresiduals(S.ARIMA)
######################Forecast
S.ARIMA.error <- valid - forecast::forecast(S.ARIMA, h = 168)$mean

# Calculate prediction error statistics (MAE and MAPE)
S.ARIMA.MAE <- mean(abs(S.ARIMA.error))
S.ARIMA.MAPE <- mean(abs(S.ARIMA.error)/abs(valid))*100

S.ARIMA.MAE #290.65
S.ARIMA.MAPE #36.776
#----------------------------------------------------------------
S2.ARIMA <- Arima(training,order=c(1,0,0),seasonal=c(1,1,1))
summary(S2.ARIMA)
#BIC: 266101.4
#MAPE: 2.082813
#MAE: 21.28893
checkresiduals(S2.ARIMA)

#FORECASTTTTTTTTTTTTTTTT
S2.ARIMA.error <- valid - forecast::forecast(S2.ARIMA, h = 168)$mean

# Calculate prediction error statistics (MAE and MAPE)
S2.ARIMA.MAE <- mean(abs(S2.ARIMA.error))
S2.ARIMA.MAPE <- mean(abs(S2.ARIMA.error)/abs(valid))*100

S2.ARIMA.MAE #226.28
S2.ARIMA.MAPE #27.386
#----------------------------------------------------------------

S3.ARIMA <- Arima(training,order=c(1,0,1),seasonal=c(0,1,2))
summary(S3.ARIMA)
#BIC: 260349
#MAPE: 1.85602
#MAE: 18.90246
checkresiduals(S3.ARIMA)
#FORECASTTTTTTTTTTTTTTT
S3.ARIMA.error <- valid - forecast::forecast(S3.ARIMA, h = 168)$mean

# Calculate prediction error statistics (MAE and MAPE)
S3.ARIMA.MAE <- mean(abs(S3.ARIMA.error))
S3.ARIMA.MAPE <- mean(abs(S3.ARIMA.error)/abs(valid))*100

S3.ARIMA.MAE #261.6598
S3.ARIMA.MAPE #31.55

#==----------------------------------------
S4.ARIMA <- Arima(training,order=c(0,0,1),seasonal=c(0,1,2))
summary(S4.ARIMA)
#BIC: 260349
#MAPE: 1.85602
#MAE: 18.90246
checkresiduals(S4.ARIMA)

#FORECASTTTTTTTTTTTTTTTTTTTT
S4.ARIMA.error <- valid - forecast::forecast(S4.ARIMA, h = 168)$mean

# Calculate prediction error statistics (MAE and MAPE)
S4.ARIMA.MAE <- mean(abs(S4.ARIMA.error))
S4.ARIMA.MAPE <- mean(abs(S4.ARIMA.error)/abs(valid))*100

S4.ARIMA.MAE #291.21
S4.ARIMA.MAPE #35.9946
```


```{r x}
#==----------------------------------------
S5.ARIMA <- Arima(training,order=c(3,1,2),seasonal=c(0,1,2))
summary(S5.ARIMA)
#BIC: 
#MAPE:
#MAE: 
checkresiduals(S5.ARIMA)

#FORECAST
S5.ARIMA.error <- valid - forecast::forecast(S5.ARIMA, h = 168)$mean

# Calculate prediction error statistics (MAE and MAPE)
S5.ARIMA.MAE <- mean(abs(S5.ARIMA.error))
S5.ARIMA.MAPE <- mean(abs(S5.ARIMA.error)/abs(valid))*100

S5.ARIMA.MAE #162.475
S5.ARIMA.MAPE #19.98%
```