seasonaldecompositiontry.Rmd

---
title: "R Notebook"
output: html_notebook
---

This is an [R Markdown](http://rmarkdown.rstudio.com) Notebook. When you execute code within the notebook, the results appear beneath the code. 

Try executing this chunk by clicking the *Run* button within the chunk or by placing your cursor inside it and pressing *Ctrl+Shift+Enter*. 

```{r}
load_history <- read.csv("E:/ETUDE/EP/MAP573/Projet/global-energy-forecasting-competition-2012-load-forecasting/Load_history.csv", header=TRUE)
library(naniar)
library(VIM)
library(FactoMineR)
library(missMDA)
load_history<-load_history %>% dplyr::na_if('')
#res<-summary(aggr(load_history, sortVar=TRUE))$combinations
matrixplot(load_history, sortby = 2)
```


```{r}
gg_miss_var(load_history)
```

```{r}
res<-summary(aggr(load_history, sortVar=TRUE))$combinations
```


```{r}
library(dplyr)
varnames <- paste("data_",1:20,sep="")
for (i in 1:20){
  assign(varnames[i],value=filter(load_history,zone_id == i))
}
```


```{r}
matrixplot(data_1, sortby = 2)
```


```{r}
library(forecast)
?snaive
```

```{r}
library(fpp2)
rm(ausbeer)
data(ausbeer)
#show(ausbeer)
forecast(ausbeer)
```

```{r}

beer2 <- window(ausbeer,start=1992,end=c(2007,4))

autoplot(beer2) +
  autolayer(meanf(beer2, h=11)$mean, series="Mean") +
  autolayer(naive(beer2, h=11)$mean, series="Naïve") +
  autolayer(snaive(beer2, h=11)$mean, series="seasonal naïve") +
  autolayer(forecast(beer2, h=11)$mean,series="Forecast")+
  ggtitle("prdiction") +
  xlab("year") + ylab("ML") +
  guides(colour=guide_legend(title="prediction")) +
  theme(text=element_text(family="STHeiti"))+
  theme(plot.title = element_text(hjust = 0.5))
```

```{r}

beer2 %>%
  stl(t.window=13, s.window="periodic", robust=TRUE) %>%
  autoplot()
```

```{r}
autoplot(beer2)
```


```{r}
fit <- stl(beer2,t.window=13, s.window="periodic", robust=TRUE)

fit %>% forecast(method="arima") %>% autoplot() 
```


```{r}
dim(data_1)
D_1<-data_1[,5:28]
D_1_1<-as.vector(t(D_1))
#Data_1ezone_dim1

```

```{r}
dim(data_1)
D_1<-data_1[,5:28]
D_1_1t<-as.vector(t(D_1))
D_1_1 <- c(length(D_1_1t))
library(stringr)
for (i in 1:length(D_1_1t)){
  D_1_1t[i] <- str_remove(D_1_1t[i],",")
  D_1_1[i] <- strtoi(D_1_1t[i])
}
D_1_1ts<-ts(D_1_1,frequency=24) #time sequence
```



data for each hour
```{r}
D_1_1h<-c(dim(D_1)[1])
for (i in 1:dim(D_1)[1]){
  D_1_1h[i]<-strtoi(str_remove(D_1[i,1],","))
}
D_1_1hts<-ts(D_1_1h,frequency=365)
```


```{r}
Day <- 30
ts(D_1_1[1:(24*Day)],frequency=24) %>%
  stl(t.window=7, s.window="periodic", robust=TRUE) %>%
  autoplot()

#stl(ts(D_1_1[1:(24*Day)],frequency=24),t.window=7, s.window="periodic", robust=TRUE)$time.series[,2]


fit <- stl(ts(D_1_1[1:(24*Day)],frequency=24),t.window=7,s.window="periodic",robust=TRUE)
#fit$time.series[,2]

fit %>% forecast(method="naive",h=(24*10)) %>% autoplot() 
fit %>% forecast(method="arima",h=(24*10)) %>% autoplot() 
```
```{r}
#download.packages("seasonal")
library(seasonal)
Day<-70
autoplot(seas(ts(D_1_1[1:(24*Day)],frequency=24,start = 2004)))
```

Add a new chunk by clicking the *Insert Chunk* button on the toolbar or by pressing *Ctrl+Alt+I*.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the *Preview* button or press *Ctrl+Shift+K* to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike *Knit*, *Preview* does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.