covid19_confirmed_cases_kerala.Rmd

---
title: covid19_confirmed_cases_in_kerala
output: html_document
By prakruthi jain YS
   

Load dataset
```{r}
##loading dataset
data=read.csv("Covid 19 Confirmed Cases-Kerala.csv")
```

head() :shows top few rows in dataset by default it shows the first 6 rows
```{r}
##to get first few row
head(data)
```

shows top 3 rows in the dataset 
```{r}
##to get first 3 row
head(data, n=3)
```

to extract data of particular column 
```{r}
head(data$Confirmed)
data$Confirmed[3]
data$Confirmed[1:6]
```

tail() :shows last few rows in dataset by default it shows the last 6 rows
```{r}
##to get last few row
tail(data)
```

displays selected rows and columns
```{r}
##displays 3rd row and 2nd column
data[3,2]
```

as.matrix() converts dataframe into matrix
```{r}
head(as.matrix(data))
```

rbind() and cbind() create matrices by combining several vectors of same length
```{r}
##rbind() combines as rows
head(rbind(data))
##cbind() combines vectors as column
head(cbind(data))
```

attach() function used to access the variables present in the data framework without calling the dataframe
```{r}
data1 = attach(data)
data1
```

stack() used to transform data available as separate columns in a dataframe or list into a single column
```{r}
head(stack(data))
```

sort() command sort data in order by default sorts in ascending order
```{r}
##to sort data
head(sort(data$Confirmed))
```

order( ) is same as sort() command 
```{r}
head(order(data$Confirmed))
```

length() :command used to determine the number of items in an dataset
```{r}
## to determine number of items
length(data)
```

dimnames() gives names of both rows and column
```{r}
##to see row and column names
dimnames(data)
```

class() :command used to see type of data
```{r}
##class() to obtain info about the type of object
class(dimnames(data))
class(data)
class(data$Date)
class(data$Confirmed)
```

str() command shows structure of data
```{r}
##structure of dataset
str(data)
```

structure returns the data with further attributes set
```{r}
head(structure(data))
```

summary() command used to get information about the data
```{r}
##summary of dataset
summary(data)
summary(data$Confirmed)
```

statistics of data(mean,median,max,min,mode,standard deviation, variance, mad)
```{r}
##mean of confirmed
mean(data$Confirmed)
##median of confirmed
median(data$Confirmed)
##maximum number in confirmed column
max(data$Confirmed)
##mode of confirmed data
mode(data$Confirmed)
##minimum value in confirmed data
min(data$Confirmed)
##standard deviation of confirmed data
sd(data$Confirmed)
##variance in confirmed data
var(data$Confirmed)
##median absolute deviation of confirmed data
mad(data$Confirmed)
```

quantile of confirmed data with 25th, 50th, 75th percentile
```{r}
##quantile 25%,50%,75%
quantile(data$Confirmed, 0.25)
quantile(data$Confirmed, 0.5)
quantile(data$Confirmed, 0.75)
quantile(data$Confirmed, c(0.25,0.5,0.75))
```

returns five number summary of confirmed data(min, lower-hinge,median,upper-hinge, max)
```{r}
#five number of data
fivenum(data$Confirmed)
```

ls.str() displays the internal structure of R object
```{r}
## all the named objects at once using ls.str() command
ls.str(data)
ls.str(pattern = "data")
```

rank() command gives rank number like order() command with slight difference. when the values are same the ranks are shared between them but order() doesnt do this
```{r}
##rank of data
rank(data$date)
rank(data$confirmed)
```

displays first column
```{r}
##to display first column
head(data[1])
```

cummulative commands
```{r}
##cummulative sum
head(cumsum(data$Confirmed))
##cumulative max
head(cummax(data$Confirmed))
##cummulative min
head(cummin(data$Confirmed))
##cummulative product
head(cumprod(data$Confirmed))
```

install.packages() to install packages from repositories
```{r}
install.packages("dplyr", repos = "https://cran.us.r-project.org")
install.packages("magrittr",repos = "https://cran.us.r-project.org")
install.packages("ggplot2", repos = "https://cran.us.r-project.org")
```

installed.packages() to see packages which are installed
```{r}
##to see installed packages
head(installed.packages())
```

search() command to see all the packages available 
```{r}
##to see all the packages
search()
```
 
library() command to load the packages
```{r}
##to Load the packages
library(dplyr)
library(magrittr)
library(ggplot2)
```

mutate() to add new column for existing dataset
```{r}
##to add new column to dataset with values NA
head(data %>% mutate(necol = NA))
```

select(-x) command selects the columns based on condition
```{r}

##select() command to select columns removing confirmed column
head(select(data, -Confirmed))
head(data %>% select(-Confirmed))
```

names() command to view name of column and row
```{r}
##row names and column named
rownames(data)
colnames(data)
names(data)
```

rename() command to change the name of the column
```{r}
##to make a copy of dataset and rename column name
temp = data
head(temp %>% rename(Confirm = Confirmed))
```

to change the column position 
```{r}
##to bring last column to first
head(data %>% select(Confirmed,Date))
head(data %>% select(Confirmed,everything()))
```

table() command  gives number of time the values repeated
```{r}
##to see how many no of times the value repeated
table(data$Confirmed)
head(as.table(data$Confirmed))
```

to check the data is in dataframe or not
```{r}
is.data.frame(data)
is.table(data)
```

seq_along() function creates vector that contains a sequence of numbers from 1 to the length of the object
```{r}
seq_along(data)
```

stem() command extracts the numeric data and splits them into two parts namely, the stem and leaf
```{r}
stem(data$Confirmed)
##increasing the number of bins by adding a scale = 2 instruction
stem(data$Confirmed, scale = 2)
```

filter() gives rows where conditions are true
```{r}
##filter() command to get data when confirmed is 0
filter(data,Confirmed == 0)
```

to calculate new column based on existing column
```{r}
##add new column named "meanconfirmed" which is calculated based on existing "confirmed" column
head(data %>% mutate(meanconfirmed = Confirmed-mean(Confirmed )))
```

to plot dotplot with ggplot 
```{r}
##the covid cases confirmed on particular dates  is plotted with pointgraph (date vs confirmed)
ggplot(data = data, aes(x = Date, y = Confirmed))+geom_point()+labs(x = "date", y = "confirmed cases", title = " dotgraph of covid19 cases in kerala", subtitle = "from 31/01/2020 to 10/08/2021")
```

to plot bargraph with ggplot
```{r}
##to plot bargraph of covid cases confirmed on particular dates 
ggplot(data = data, aes(x = Date, y = Confirmed))+geom_bar(stat = "identity", fill = "purple")+labs(x = "date", y = "confirmed cases", title = "covid19 cases in kerala", subtitle = "from 31/01/2020 to 10/08/2021")
```

plot() function to create a graphical representation of some data
```{r}
##plots line 
plot(data$Confirmed, type = 'l', col = 'red')
##plots point
plot(data$Confirmed, type = "p")
##plots both line and point
plot(data$Confirmed, type = 'b', axes = TRUE, ylab = 'Confirmed cases')
##to plot overplotted points and lines
plot(data$Confirmed, type = "o")
##to plot stair steps
plot(data$Confirmed, type = "s")
##to plot histogram
plot(data$Confirmed, type = "h")
```

barplot() plots bargraph of confirmed data with date
```{r}
##bar plot
barplot(table(data$Confirmed), xlab = 'date', ylab = 'confirmed cases')
barplot(data$Confirmed, names = data$Date, xlab = 'date', ylab = 'confirmed cases')
```

installing lubridate package that makes it easier to work with dates. load that package
```{r}
## install lubridate package 
install.packages("lubridate", repos = "https://cran.us.r-project.org")
library(lubridate)
```

using mutate() add new columns "year" and "month" by separating year and month form date(year-month-date)
using group_by() group the data by months and year 
using summarise() create new dataframe with fewer rows
```{r}
##to separate date as month and year 
new_data= data %>% mutate(year = year(Date), month = month(Date)) %>% group_by(year, month) %>%  summarise(mean_var = mean(Confirmed))
```

filter() filters the data and gives the data of confirmed cases from date "25-7-2021" to "1-8-2021"
```{r}
##filter date and get data between 1aug2021 and 10aug2021
fildata = data %>% filter(Date >= as.Date("2021-07-25") & Date <= as.Date("2021-08-01"))
```

pie() to plot piechart of confirmed cases for filtered data(i.e from "25-7-2021" to "1-8-2021" )
```{r}
##pie graph
pie(fildata$Confirmed, labels = fildata$Date)
```

as.Date() function used to change the format of date
```{r}
##create new data>frame with the newly formatted date field
head(data %>% mutate(Date = as.Date(Date , format = "%Y-%m-%d")))
```

hist() computes histogram of the given datavalues
geom_histogram() plots histogram using ggplot. 
```{r}
##plot histogram
hist(data$Confirmed)
##histogram is plotted with the function hist() and the option freq = False plots probability densities instead of frequencies
hist(data$Confirmed, col = "blue", xlab = "confirmed cases", ylim = c(0, 0.002), freq = FALSE)
##using ggplot histogram can be plotted using geom_histogram() function
ggplot(data, aes(x = Confirmed)) + geom_histogram()
```

calculating density using density() function
```{r}
density(data$Confirmed)
```
```{r}
##to plot histogram
hist(data$Confirmed, freq = FALSE, main = "Histogram and density")
##add density to plot
lines(density(data$Confirmed), lwd = 2, col ="red")
##plots the density without histogram
plot(density(Confirmed), lwd = 2, col = "red", main = "Density")
```

rnorm() function simulates random variates having a specified normal distribution
```{r}
##with mean =0 and standarddeviation = 1
head(rnorm(data$Confirmed, mean = 0 ,sd = 1))
```

pnorm() function gives the cumulative disttribution function(CDF) of the normal distribution 
```{r}
head(pnorm(data$Confirmed, mean = 0 , sd = 1))
```

qnorm() function calculates the inverse of CDF of the normal distribution
```{r}
head(qnorm(data$Confirmed, mean = 0 , sd = 1))
```

dnorm() function calculates the probability density function of the normal distribution
```{r}
head(dnorm(data$Confirmed, mean = 0 , sd = 1))
```

normal Q-Q plot qqnorm() is a generic function which produces a normal qq plot
```{r}
qqnorm(Confirmed)
qqnorm(Confirmed, main = "qq plot of Confirmed", xlab = "theoretical quantiles", ylab = "quantiles for Confirmed")
qqline(data$Confirmed, lwd = 2, lty = 6)
```

t.test() function performs t-tests. t-test used to determine whether the means of two groups are equal to eachother
```{r}
##mean specified by the null hypothesis mu = 0
t.test(Confirmed, mu =5)
t.test(Confirmed, mu =5, alternative = "greater")
```

chisq.test() function used to perform chi-square test
```{r}
chisq.test(Confirmed)
```

there are two types of barcharts geom_bar() and geom_col() . geom_bar() makes the height of the bar proportional to the number of cases in each group, geom_col() to represent values in the data
```{r}
##basic bar graph
library(gcookbook)
ggplot(fildata, aes(x = Date, y = Confirmed))+geom_col()
ggplot(fildata, aes(x = Date, y = Confirmed))+geom_col(fill = "lightpink", colour = "blue")
ggplot(fildata, aes(x = Date)) + geom_bar(width = 0.3, position = position_dodge(0.5))
```


filter data by date from 2020-03-09 to 2020-03-20 and the plot bargraph using geom_col in ggplot 
```{r}
newdata = data %>% filter(Date >= as.Date("2020-03-09") & Date <= as.Date("2020-03-20"))
ggplot(newdata, aes(x= Date, y=Confirmed)) + geom_col(position = 'fill') + scale_fill_brewer(palette = "Pastel1")

```

group the data by date and add new column named "percent_confirmed" which is percentage og comfirmed data and plot bargraph date vs percent_confirmed
```{r}
## grouping data by Date and calculating percentage of confirmed cases and add that to existing dataset as percent_confirmed using mutate 
head(data %>% group_by(Date))
new_data = data %>% group_by(Date) %>% mutate( percent_Confirmed = Confirmed / sum(Confirmed) * 100)
ggplot(new_data , aes(x = Date, y = percent_Confirmed))+ geom_col()
```

new columns "year" and "month" is added ang grouping that data by moth and year
```{r}
##seperate date into month and year
data = data %>% mutate(year = year(Date), month = month(Date)) %>% group_by(year, month)
##stem with conditional statement
with(data, stem(data$Confirmed[data$year == "2020"]))
```

plotting bargraph month vs confirmed and adding labels
```{r}
##adding labels to a bar graph
ggplot(data, aes(x= month, y=Confirmed)) + geom_col() + geom_text(aes(label = Confirmed), vjust = -0.2)
ggplot(data, aes(x= month, y=Confirmed)) + geom_col() + geom_text(aes(label = Confirmed), vjust = 1.5, colour = "white")
```

to plot boxplots
```{r}
## to plot boxplot
boxplot(data$Confirmed ~ data$year, data = data)
boxplot(data$month)
boxplot(data$Confirmed ~ data$month)
boxplot(data$Confirmed ~ data$year, horizontal = T)
```


plotting line graph in ggplot using geom_line() 
```{r}
##to plot linegraph
ggplot(data, aes(x = month, y = Confirmed))+geom_line(aes(color = year))+labs(x = "date", y = "confirmed cases")
```

```{r}
##line graph with points
ggplot(data, aes(x = month, y= Confirmed)) + geom_area(fill = "blue", alpha = .2) + geom_line()
ggplot(data, aes(x = month, y = Confirmed)) + geom_line() + geom_point()
```

```{r}
##making a line graph with multiple lines(confirmed cases for same month but in 2 different year 2020 and 2021)
ggplot(data, aes(x = month, y = Confirmed, colour = year)) + geom_line()
```

```{r}
##line graph with different colors
ggplot(data, aes(x = month, y = Confirmed, fill = year)) + geom_line() + geom_point(size = 4, shape = 21)
```

```{r}
##line graph with dashed line
ggplot(data, aes(x = month, y = Confirmed)) + geom_line(linetype = "dashed", size =1, colour = "purple") 
```

```{r}
##line graph with dotted line
ggplot(data, aes(x = month, y = Confirmed)) + geom_line(linetype = "dotted", size =1, colour = "purple")
```

```{r}
##line graph with point custom size, shape, colour, fill
ggplot(data, aes(x = month, y = Confirmed)) + geom_line() + geom_point(size = 4, shape = 22, fill = "pink", colour = "darkred")
```

```{r}
##points with white fill
ggplot(data, aes(x = month, y = Confirmed)) + geom_line(aes(color = year)) + geom_point(size = 4, shape = 21 , fill = "white")
```

```{r}
ggplot(data, aes(x = month, y = Confirmed)) + geom_line(aes(color = year)) + geom_point() +scale_y_log10()
```


```{r}
ggplot(data, aes(x= year, y=Confirmed)) + geom_col(aes(color = month)) + geom_text(aes(label = Confirmed), vjust = -0.2)
ggplot(data, aes(x= month, y=Confirmed)) + geom_col(aes(color = year)) + geom_text(aes(label = Confirmed),vjust = 1.5, colour = "white")
ggplot(data , aes(x = year, y = Confirmed))+ geom_col(aes(color = year))
ggplot(data, aes(x = year)) + geom_bar(width = 0.3, position = position_dodge(0.5))
```
 plotting scatterplot
```{r}
##scatter plot
plot(data$month, data$Confirmed, pch = 25)
```

to get data of confirmed cases in the year 2020 of all the month dated as 01
```{r}
## assign each month to the same year and day for plotting
as.Date(paste0("2020-", data$month, "-01"), "%Y-%m-%d")
```

plotting bargraph and using facet_wrap() function which partitions a plot into matrix of panels
```{r}
##plots confirmed cases in whole year 
data %>% ggplot(aes(x = year, y= Confirmed)) + geom_bar(stat = "identity", fill = "darkorchid4") + facet_wrap(~ year, ncol = 3) + labs(title = "yearly report", y = "confirmed cases", x= "year") + theme_bw(base_size = 15)

##plots confirmed cases in all months of the year 2020 and 2021
data %>% ggplot(aes(x = year, y= Confirmed)) + geom_bar(stat = "identity", fill = "darkorchid4") + facet_wrap(~ month, ncol = 3) + labs(title = "monthly report for different year", y = "confirmed cases", x= "year") + theme_bw(base_size = 15)

##plots confirmed cases in all the months
data %>% ggplot(aes(x = month, y= Confirmed)) + geom_bar(stat = "identity", fill = "darkorchid4") + facet_wrap(~ month, ncol = 3) + labs(title = "report of each month", y = "confirmed cases", x= "year") + theme_bw(base_size = 15)

data %>% ggplot(aes(x = month, y= Confirmed)) + geom_bar(stat = "identity", fill = "darkorchid4") + facet_wrap(~ year, ncol = 3) + labs(title = "yearly report", y = "confirmed cases", x= "year") + theme_bw(base_size = 15)
```


```{r}
data %>%  mutate(month2 = as.Date(paste0("2020-", month, "-01"), "%Y-%m-%d")) %>%  ggplot(aes(x = month2, y = Confirmed)) + geom_bar(stat = "identity", fill = "darkblue") + facet_wrap( ~ year, ncol = 3) + labs(title = "monthly report", subtite = "data plotted by year", y = "confirmed cases", x = "year") + theme_bw(base_size = 15) + scale_x_date(date_labels = "%b")
```

plotting point graph
```{r}
##plot the data using ggplot2,pipes and facet_wrap
data  %>% ggplot(aes(x= year, y = Confirmed)) + geom_point(color = "darkorchid4") + facet_wrap( ~ month, ncol = 3) + labs(title = "monthly report", subtite = "use facets to plot by a variable - year", y = "confirmed cases", x = "year") + theme_bw(base_size = 15) 
```

```{r}
##plot pointplot the data using ggplot2 and pipes
data %>% ggplot(aes(x = month, y = Confirmed)) + geom_point(color = "darkorchid4") + labs(title = "confirmed cases in 2020 and 2021", subtitle = "The data frame is sent to the plot using pipes", y = "confirmed cases", x = "month") + theme_bw(base_size = 15)

ggplot(data, aes(x = month, y = Confirmed)) + geom_point(color = "darkorchid4") + labs(title = "confirmed cases in 2020 and 2021", subtitle = "note using pipes", y = "confirmed cases", x = "month") + theme_bw(base_size = 15)
```

plot of confirmed cases from 9th march to 20th march
```{r}
##to plot bargraph of confirmed cases from 9th march to 20th march
newdata = data %>% filter(Date >= as.Date("2020-03-09") & Date <= as.Date("2020-03-20"))
newdata %>% ggplot(aes(x = Date, y = Confirmed)) + geom_bar(stat = "identity", fill = "darkorchid4") + facet_wrap( ~year, ncol = 3) + labs( title = "confirmed cases" , subtitle = "data plotted by year",y = "confirmed cases", x = "date") + theme_bw(base_size = 15)
```

add new column named "month2" which has dates of 2020year and date 01 , and then group the data by month2 and see the summary of confirmed
```{r}
##calculate the sum confirmed cases for each month in 2020 and date =01
data %>%  mutate(month2 = as.Date(paste0("2020-", month, "-01"), "%Y-%m-%d")) %>% group_by(month2) %>% summarise(sum_confirmed = sum(Confirmed))
```

month and year is grouped and summarise() creates bew variable named "max_conf" 
```{r}
## calculate the sum confirmed 
data %>% group_by(month, year) %>% summarise(max_conf = sum(Confirmed))
```

with() function evaluates an R expression in an environment constructed based on a dataframe
```{r}
with(data, (data$Confirmed[data$year == "2020"]))
```

pairs() function is used to return a plot matrix, consisting of scatter plots 
```{r}
pairs(~ Confirmed + month , data = data)
```

dotchart() function ismused to create a dot chart of the specified data.A dot chart is defined as a plot which is used to draw a cleveland dot plot.
```{r}
dotchart(data$Confirmed)
```
aov() function is used to do ANOVA in R . na.omit() function removes all incomplete cases of a dataobject 
```{r}
data.aov = na.omit(data$Confirmed)
data.aov
summary(data.aov)
```

frequency polygon geom_freqpoly displays the counts with line , binwidth= the width of the bins
```{r}
data %>% ggplot(aes(year)) + geom_freqpoly(binwidth = 0.5)
data %>% ggplot(aes(month)) + geom_freqpoly(binwidth = 0.1) + scale_y_sqrt()
```

factor() is used to categorize and store the data having limited number of differnt values. it stores data as a vector of integer valuesre
```{r}
factor(data$Confirmed)
factor(data$year)
factor(data$Date)
factor(data$month)
```

rep() replicates the element of lists and vector
```{r}
head(rep(data))
head(rep(data$Confirmed))
```

col() function gets the column number of a matrix
```{r}
col(data, as.factor = FALSE)
y <- matrix(rep(1:9),3,3)
col(y)
```

to get all the row values of 2nd and 3rd column i.e., confirmed and year 
```{r}
x = data[ , 2:3]
head(x)
```

colMeans() function gives mean value of the columns in the dataset
rowMeans() function gives mean value of the rows in the dataset
colSums() and rowSums() gives the sums of matrix or array columns
```{r}
colMeans(x)
rowMeans(x)
colSums(x)
rowSums(x)
```
 apply() function used to apply a function across an array, matrix or dataframe
```{r}
##vmean on columns
apply(data[ , 2:3], 2 , mean)
##mean on rows
apply(data[ , 2:3], 1 , mean)
##median
apply(data[ , 2:3], 2 , median)
##variance
apply(data[ , 2:3], 2 , var)
```

tapply() to summarize using a grouping variable
```{r}
tapply(data$Confirmed, data$year, FUN = mean)
tapply(data$Confirmed, data$year, FUN = var)
tapply(data$Confirmed, data$year, FUN = sd)
tapply(data$Confirmed, data$year, FUN = median)

```

na.omit() function removes all incomplete cases of a dataobject 
```{r}
head(na.omit(data))
```
 object( ) and ls() functions both does same operation
 object() returns a list of objectnames
 ls() function is used to list the names of all the objects that are present in working library
```{r}
objects(data)
ls(data)
```

Simple linear Regression
lm() function can be used to determine the beta coefficients of liner model
```{r}
data_regr = lm(month ~ Confirmed, data)
data_regr
##summary
summary(data_regr)
##confidence interval
confint(data_regr)
cor.test( ~ data$Confirmed + data$month)
coef(data_regr)
```

fitted values
```{r}
head(fitted(data_regr))
```

Residuals
```{r}
head(residuals(data_regr))
```

Formula
```{r}
formula(data_regr)
formula(data)
data_regr$call
```

using segments() command for Error Bars
```{r}
meandata = apply(data[ , 2:3], 2, mean)
meandata
sd_data = apply(data[, 2:3], 2, sd)
sd_data
sumdata = apply(data[ , 2:3], 2, sum)
sumdata
df = sumdata/meandata
df
meandata + df
max(meandata + df)
maxdata = round(max(meandata + df) + 0.5 , 0)
maxdata
plot = barplot(meandata , ylim = c(0 , maxdata))
plot
segments(plot , meandata + df, plot , meandata - df)
segments(plot - 0.2, meandata - df, plot + 0.2, meandata - df)
box()
title(ylab = "count")
```

Creating mathematical expressions
```{r}
plot(1:10 , 1:10, type = 'n')
opt = par(cex = 1.5)
text(1 , 1, expression(hat(x)))
text(2 , 2, expression(alpha == x))
text(3 , 3, expression(beta == y))
text(4 , 4, expression(frac(x, y)))
text(5 , 5, expression(sum(x)))
text(6 , 6, expression(sum(x^2)))
text(7 , 7, expression(bar(x) == sum(frac(x[i], n), i==1, n)))
text(8 , 8, expression(sqrt(x)))
text(9 , 9, expression(sqrt(x , 3)))
```

aggregate() function splits the data into subsets and computes summary statistics for each subsets and returns the result in agroup by form
```{r}
aggregate(data[ , 2:3], by = list(data$Confirmed), FUN = mean)
aggregate(data$Confirmed ~ data$month, FUN = mean)
aggregate(data$Confirmed ~ data$month * data$year, FUN = mean)
aggregate(cbind(data$Confirmed, data$month), by = list(data$year), FUN = mean)
```