prediction.Rmd

---
title: "Prediction-fatemeh nosrat"
author: "Fatemeh Nosrat"
date: "12/13/2018"
output: html_document
---

```{r}
---
title: "Project2"
author: "Fatemeh Nosrat"
date: "12/12/2018"
output:
  word_document: default
  html_document: default
  pdf_document: default
---

```{r}
#data in item_categories.csv
item_categories <- read.csv(file="/Users/fatemehnosrat/Desktop/Applied Regression/all-2/item_categories.csv",header = TRUE, na.strings=c("",".","NA"))

# a summary of item_categories
summary(item_categories)

# data in saless_train.csv
sales_train <- read.csv(file="/Users/fatemehnosrat/Desktop/Applied Regression/all-2/sales_train.csv",header = TRUE, na.strings=c("",".","NA"))

# a summary of sales_train
summary(sales_train)

#data in shops.csv
shops <- read.csv(file="/Users/fatemehnosrat/Desktop/Applied Regression/all-2/shops.csv",header = TRUE, na.strings=c("",".","NA"))

# a summary of shops.csv
summary(shops)
 
#data in items.csv
items <- read.csv(file="/Users/fatemehnosrat/Desktop/Applied Regression/all-2/items.csv",header = TRUE, na.strings=c("",".","NA"))

# a summary od shops
summary(items)
#data in test.csv
test <- read.csv(file="/Users/fatemehnosrat/Desktop/Applied Regression/all-2/test.csv",header = TRUE, na.strings=c("",".","NA"))

# a summary of test.csv
summary(test)

# loading plyr library, we need plyr to use "join"
library(plyr)

 #join 2 data frames (sales_train and items) by their common column (item_id) 
combined_data <- join (x = test, y = sales_train, by = c("item_id"))

head(combined_data)

# a summary of combined_data
summary(combined_data)

# load lubricate, we need this to work with "time and date"
library(lubricate)
#column date in combined_data
combined_data$date <-dmy(combined_data$date)

#column year in combined_data
combined_data$year<-year(combined_data$date)

#column month in combined_data
combined_data$month<-month(combined_data$date)

#column day in combined_data
combined_data$day<-month(combined_data$day)

#column weekday in combined_data
combined_data$weekday<-weekdays(combined_data$date)

#convert column year in combined_data to a factor
combined_data$year <- as.factor(combined_data$year)

# convert column month in combined_data  to a factor
combined_data$month <- as.factor(combined_data$month)

#convert column weekday in combined_data to a factor
combined_data$weekday <- as.factor(combined_data$weekday)

#convert column shop_id to a factor 
combined_data$shop_id <- as.factor(combined_data$shop_id)

#convert column item_id to a factor
combined_data$item_id <- as.factor(combined_data$item_id)

#convert item_category_id to a factor
combined_data$item_category_id <- as.factor(combined_data$item_category_id)

##Explanation of aggregate:
#The process involves two stages. First, collate individual cases of raw data together with a grouping variable.
#Second, perform which calculation you want on each group of cases. These two stages are wrapped into a single function.

#calculating the total items by month
aggregate(item_cnt_day~month, combined_data, sum)

#loading data.table for creating a data table

library(data.table) 

#use function as.dara.table for combining data
train_datatable = as.data.table(combined_data)

summarized_month_shop_item = train_datatable[, list(item_cnt_month=(sum(item_cnt_day))/12), by = c("date_block_num", "month","shop_id", "item_category_id", "item_id", "item_price")]

# a summary of summarized_month_shop_item
summary(summarized_month_shop_item)

head(summarized_month_shop_item)


library(MASS)

#2 is added to avoid negative values for log transformation, it will be subtracted later
summarized_month_shop_item$item_cnt_month <- summarized_month_shop_item$item_cnt_month + 2

#log transformed linear regression model
sales_model = lm(formula = log(item_cnt_month) ~ date_block_num + month + shop_id + item_category_id, data = summarized_month_shop_item)

#summary of sales_model
summary(sales_model)

#Assign 11 to month for November
test$month <- 11

#convert month to a factor
test$month <- as.factor(test$month)
test$shop_id <- as.factor(test$shop_id)

#Assign 34 to date_block_num for Nov 2015
test$date_block_num <- 34
head(test)


```


```

## R Markdown

This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see <http://rmarkdown.rstudio.com>.

When you click the **Knit** button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:

```{r cars}
summary(cars)
```

## Including Plots

You can also embed plots, for example:

```{r pressure, echo=FALSE}
plot(pressure)
```

Note that the `echo = FALSE` parameter was added to the code chunk to prevent printing of the R code that generated the plot.