AnkitHw5.Rmd

---
title: "AnkitHw5"
author: "Ankit"
date: "8/27/2021"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

Loading relevant packages

```{r}
library(tidyverse)
```

Reading the dataset into a data frame

```{r}
library(readr)
hotel_bookings <- read_csv("_data/hotel_bookings.csv")
```

Overview of the dataset
```{r}
glimpse(hotel_bookings)
```

Taking a look at the first few rows
```{r}
head(hotel_bookings)
```

Are Na values present in the dataset?

```{r}
hotel_bookings %>% is.na()%>% sum()
```

There are 4 Na values in the data set.

Identifying the positions of NA values in the dataset

```{r}
hotel_bookings %>% is.na() %>% which(arr.ind = TRUE)
```

arr.ind is TRUE because we need the positions of NA values in the Dataset

Na values are present in the 11th column. So, examining the rows which have NA values

```{r}
hotel_bookings[c(40601,40668,40680,41161), ]
```

The children column has NA values for the above rows

Removing the rows which have Na values
```{r}
hotel_bookings <- hotel_bookings %>% filter(!is.na(children))
```

Checking whether data set still has Na values

```{r}
hotel_bookings %>% is.na()%>% sum()
```

```{r}
hotel_bookings %>% mutate(arrival_date_month = factor(arrival_date_month,
    levels = c("January", "February", "March", "April", "May", "June", "July",
    "August", "September", "October", "November", "December"), ordered = TRUE))
```


Frequency table of Hotels in the dataset
```{r}
hotel_bookings %>%
  group_by(hotel) %>% summarise(freq=n())
```

The number of city hotels in the data set is very large relative to the number of resort hotels.

Observing average daily rates by assigned room type in Hotels

```{r}
hotel_bookings %>% group_by(assigned_room_type, hotel) %>% select(assigned_room_type, adr, hotel) %>% summarise(mean = mean(adr))  %>% ggplot(aes(x = assigned_room_type, y = mean, color = hotel, fill= hotel)) + geom_point(shape=21, alpha=.55, size=5) + labs(title = "Average Daily Rate by Assigned Room Type in Resort Hotels", x = "Assigned Room Type", y = "Average Daily Rate")
```
Exploring the correlation between canceled hotel bookings and other variables.

```{r}
cor(hotel_bookings$is_canceled, hotel_bookings$previous_cancellations)
```

```{r}
cor(hotel_bookings$is_canceled, hotel_bookings$previous_bookings_not_canceled)
```

```{r}
cor(hotel_bookings$is_canceled, hotel_bookings$days_in_waiting_list)
```

Observing Lead Time by Customer Type in Hotels

```{r}
hotel_bookings  %>% select(customer_type, lead_time, hotel) %>% group_by(customer_type, hotel) %>% summarise(mean = round(mean(lead_time), 2)) %>% ggplot(aes(x = customer_type, y = mean, color = hotel, fill = hotel)) + 
  geom_point(shape=21, alpha=.55, size=5) + 
  geom_text(aes(label=mean,vjust=-0.3)) +
  labs(title = "Lead Time by Customer Type in Hotels", x = "Customer Type", y = "Lead time") + theme_classic() +
  theme(plot.title = element_text(face = "bold"))
```

Understanding the kinds of meals that are booked by customers.

```{r}
(a <- hotel_bookings%>% select(meal, hotel) %>%  group_by(meal, hotel) %>%  summarise(freq = n()))
ggplot(a, aes(x=meal, y=freq, fill = meal)) +
 geom_bar(stat="identity") + facet_wrap(~hotel) +
geom_text(aes(label=freq),vjust=-0.3)+
  labs(title = "Meals booked by Customers in Hotels", x = "Meal", y = "Number of meals") +
 theme_classic() +
 theme(plot.title = element_text(face = "bold"))
```