Class 1

[Link](123)

n <- 100

##making data frame

data<- [Link](
pregnant= sample(0:16,n, replace = TRUE),
glucose= pmax(round(rnorm(n, 120, 30)),50),
blood_pressure=pmax(round(rnorm(n,70,12)),40),
skin_thickness=pmax(round(rnorm(n,20,10)),0),
insulin= pmax(round(rnorm(n, 80, 40)),0),
bmi= pmax(round(rnorm(n, 30, 6),1),10),
pedigree= round(runif(n,0.1,2.5),3),
age=sample(21:64, n, replace=TRUE),
diabetes= sample(c("positive","negative"),n, replace=TRUE)
)
#sample=random pick , rnorm=normal distribution, runif=uniform decimals, round=value
rounding,

#pmax=lower bound value, pmin= upper bound

[Link](data,"C:/Users/Sayfa/Desktop/Sayfa/class/[Link]")

## after converting we use labels

a <- factor(data$diabetes,levels=c("positive","negative"),labels=c("yes","no"))
b <- cut(data$bmi, breaks = c(Inf,18.5,24.9,29.9,Inf),
labels=c("Under","Normal","Over","Obese"))
Class 2
library(readxl)
library(lubridate)
library(tseries)
data2 <- read_excel("C:/Users/Sayfa/Desktop/Sayfa/class/monthly_cloud_data.xlsx")
names(data2)
data2$Date<- ymd(data2$Date)

## Answer-1
ts_cloud<- ts(data2$Cloud_Oktas, start = c(2000,1), frequency = 12)
plot(ts_cloud, main= "Monthly Cloud Data (2000-2022)", ylab="Cloud Data", xlab="Year",
col="blue")

## Answer-2
mean(data2$Cloud_Oktas)
median(data2$Cloud_Oktas)
sd(data2$Cloud_Oktas)
hist(data2$Cloud_Oktas,main="Histogram Plot",xlab = "cloud_in_octas",ylab="frequency")

##Class Practise
BMD <- read_excel("C:/Users/Sayfa/Desktop/Sayfa/class/[Link]")
par(mfrow=c(2,2))

#Max-Temp
tmaxx <- ts(BMD$tmax, start = c(1981),frequency = 12)
plot(tmaxx, main= "Max Temp", ylab="temp", xlab="Year",col="red")

#Min-Temp
tminn <- ts(BMD$tmin, start = c(1981),frequency = 12)
plot(tminn, main= "Mini Temp", ylab="temp", xlab="Year",col="blue")

#Precipitation
rainfall <- ts(BMD$prcp, start = c(1981),frequency = 12)
plot(rainfall, main= "Rainfall", ylab="temp", xlab="Year",col="green")
Class 3
library(readxl)
monthly_cloud_data <-
read_excel("C:/Users/Sayfa/Desktop/Sayfa/class/monthly_cloud_data.xlsx")
data<- monthly_cloud_data

library(lubridate)
Date<- ymd(data$Date)

library(tidyverse)
library(dplyr)

##Filtering date

subset_data <- data%>%filter(Date>="2015-01-01" & Date<="2020-12-01")

ts_cloud<- ts(data2$Cloud_Oktas, start = c(2000,1), frequency = 12)

plot(ts_cloud, main= "Monthly Cloud Data (2000-2022)", ylab="Cloud Data", xlab="Year",
col="blue")

##PLot

ggplot(subset_data, aes(x=Date, y=Cloud_Oktas)) +

geom_line()+labs(title="Cloud Cover: 2015-2020", x="Date", y="Cloud")

##Decompose
decomposed<-decompose(ts_cloud)
plot(decomposed)

trend_part <- decomposed$trend

seasonal_part <- decomposed$seasonal
random_part <- decomposed$random
components_df <- [Link](Date = data$Date, observed=ts_cloud,
Trend=trend_part,seasonal= seasonal_part,
Random= random_part)
head(components_df)

[Link](data, "C:/Users/USER/Desktop/Sayfa/[Link]")
library(readxl)
rain_data <- read_excel("C:/Users/Sayfa/Desktop/Sayfa/class/[Link]")

library(lubridate)
library(dplyr)
library(ggplot2)

##Covert date style (separate to merge year and month)

rain_data<- rain_data%>%mutate(Date=make_date(year,mon))
tt<- ymd(rain_data$Date)
rain_data$Date<-ymd(rain_data$Date)

##RainfalPLot

ggplot(rain_data, aes(x=Date, y=prcp))+

geom_line(color="red")

##GGPLOT of Average of each year rainfall

annual_avg <- rain_data %>%

group_by(year) %>%
summarise(Average_Rainfall = mean(prcp, [Link] = TRUE))

##For each group created by group by(year), summaries() calculates the mean of prcp
(precipitation).

##Normal Plot
plot(annual_avg$year, annual_avg$Average_Rainfall, type = "o",
xlab = "Year", ylab = "Average Rainfall (mm)",
main = "Average Annual Rainfall", col = "blue")

ggplot(annual_avg, aes(x = year, y = Average_Rainfall)) +

geom_line(color = "blue") +
geom_point(color = "darkblue") +
labs(title = "Average Annual Rainfall",
x = "Year",
y = "Average Rainfall (mm)") +
theme_minimal()
Class 4
summary(mtcars$mpg) colnames(data)[c(2, 4)] <- c("cylinder",
names(mtcars) "horse-power")
head(mtcars)
data$Result<- data$`horse-power`/data$wt
#Mean names(data)
mean(mtcars$mpg)
hist(mtcars$mpg)
##Changing the level

#Median data$cylinder<- [Link](data$cylinder)

median(mtcars$mpg)
levels(data$cylinder)<-
#Mode c("low","moderate","high")
getmode<- function(v)
{
uniqv<- unique(v) ##Change to level creating new column
data$cylinder<- [Link](data$cylinder)
uniqv[[Link](tabulate(match(v,uniqv)))]
LEVEL <- c("low", "moderate", "high")
}
data$levCyl <- factor(data$cylinder, labels =
getmode(mtcars$cyl)
LEVEL)
getmode(mtcars$mpg)

##Variance and Standard Deviation #check frequency

a<- var(mtcars$mpg) table(mtcars$cyl)
b<- sqrt(a)
sd(mtcars$mpg)
##Range table
#BOX-plot
breaks <- c(-Inf, 15, 20, Inf)
boxplot(mtcars$mpg)
labels <- c("weak", "normal", "strong")
data$range_label <-
#Renaming the column cut(data$MilesPerGallon, breaks = breaks,
data<-mtcars labels = labels, right =
names(data) FALSE)
colnames(data)[1] <- "MilesPerGallon"
table(data$range_label)
Class 5
##Shapiro-Wilk normality test
[Link](mtcars$mpg)
data$am<- [Link](mtcars$am)

#W-statistic measures how close your data is to normal.

#p > 0.05 → Fail to reject the null hypothesis
#p ≤ 0.05 → Reject the null hypothesis

##T-test
[Link](mpg~am, data=mtcars)
##t = The test statistic. A more extreme value indicates
## stronger evidence against the null hypothesis.

library(readxl)
result <- read_excel("C:/Users/USER/Desktop/Sayfa/[Link]")
result$Group<-[Link](Result$Group)
[Link](Result~Group, data=result)

##Anova test
summary(aov(mpg~cyl,data=mtcars))

##Z-table
pnorm(1.96)
qnorm(0.975)
rnorm(17,3.80,0.123)

#binomial
pbinom()
rbinom(1000,5,0.5)

##Co-relation
cor(mtcars$mpg, mtcars$hp)

##Pearson's product-moment correlation

[Link](mtcars$mpg,mtcars$hp)
##Pearson's Chi-squared test
[Link](mtcars$cyl,mtcars$am)
##Regression

##linear model
model <- lm(mtcars$mpg~ mtcars$hp+mtcars$wt+mtcars$cyl, data=mtcars)
summary(model)

##Binary logistic
logit_model <- glm(mtcars$am ~ mtcars$hp + mtcars$mpg,
data = mtcars, family = binomial)
summary(logit_model)

##PLOT
par(mfrow=c(2,2))
plot(model)

Class 6
library(readxl)
BMD_data <- read_excel("C:/Users/Sayfa/Desktop/Sayfa/class/[Link]")

rain_ts<-ts(BMD_data$prcp, start = c(1981,1), frequency=12)

rain_ts
plot(rain_ts, main="Monthly Avg Rainfall",xlab="Year",ylab="Rainfall(mm)", col="brown")

## Dickey-Fuller Test
library(tseries)
[Link](rain_ts)

##1st Difference
rain_diff <- diff(rain_ts)
plot(rain_diff,main="First-order Difference")
[Link](rain_diff)

##2nd Difference
rain_diff2 <- diff(diff(rain_ts))
plot(rain_diff2, main = "Second-order Difference")
[Link](rain_diff2)

##Arima(Auto Regressive Integrated Moving Average) Model

library(forecast)
model_arima <- [Link](rain_ts)
summary(model_arima)

forecast_arima <- forecast(model_arima,h=24)

plot(forecast_arima)

Class 7
##SARIMA Model
##The forecast data is within 95% confidence interval

library(tseries)
library(forecast)

model_sarima <- [Link](rain_ts, seasonal = TRUE)

summary(model_sarima)
forecast_sarima <- forecast(model_sarima, h=60)
plot(forecast_sarima)

## If model of arima is given: order = c(p, d, q)

fit_sarima <- Arima(rain_ts, order = c(0,0,1), seasonal = c(1,1,0))
summary(fit_sarima)

##Model digonasis
checkresiduals(fit_sarima)

##No auto-correlation till 24 lag time. Statistically significant model.

spectrum(rain_ts) ## To identify dominant cycles or periodic patterns

## in a time series by analyzing frequency components.
spectrum(rain_ts,log = "no")

##Wavelet Analysis
library(WaveletComp)
data <- [Link](time=time(rain_ts), value=rain_ts)
wt <- [Link](data, [Link] = "value", [Link] = 0)
[Link](wt)
library(readxl)
rain <- read_excel("C:/Users/USER/Desktop/Sayfa/[Link]")
temp <- read_excel("C:/Users/USER/Desktop/Sayfa/[Link]")
combine <- merge( rain, temp, by = c("year"))
write_xlsx(combine,"C:/Users/USER/Desktop/Sayfa/[Link]" )

Class 8
#Making Data frame (Table)
Number <- 1:8
Diet <- c("Poor","Poor","Good","Good","Poor","Good","Poor","Good")
Sex <- c("M","F","M","F","F","F","M","M")
Weight <- c(156,180,167,190,177,183,154,160)
Morph <- c("Winged","Winged","Wingless","Intermediate",
"Winged","Intermediate","Winged","Wingless")
data <- [Link](Number, Diet, Sex, Weight, Morph)

##Add column
[Link] <- c(34,43,40,43,40,43,33,38)
data <- cbind(data, [Link])

##Add row
Newcommer <- c (9,"Poor","F",150,"Winged", 40)
data <- rbind(data, Newcommer)

##delete - array method

data3 <- data[ ,-4]
##deleting multiple column
data4 <- data[, -c(2, 4, 5)]

##Seprate data- subset

data2 <- subset(data, select = Weight)
##Seprate multiple data- subset
data2 <- subset(data, select = c(Weight, Sex))

##Changing column names

names(data$Diet) <- c("Dieeeeeet")

save(list = "data", file = "C:/Users/USER/Desktop/Sayfa/[Link]")

table(Sex)
class_data <- tibble(
name = c("X","Y","Z"), gender = c("M","F","F"), age = c(20,25,30),
Married = c(TRUE, FALSE, TRUE)
)
class_data

class_data2 <- tribble(

~name, ~sex, ~age,
"a", "F",30,
"b","M",45)
)

sayfa <- iris

library(tidyverse)
airquality <- [Link](airquality)

names(airquality)
small_dat <- select(airquality, Ozone:Wind)
head(small_dat)[1:4, ]

##Mean in monthly basis

summarize(month, oz=mean(Ozone, [Link]=T), solar=mean(Solar.R, [Link]=T))
mean(airquality$Ozone)
mean(airquality$Ozone)

Class 9
data <- airquality
View(data)

##corelation with missing data

cor(data$Ozone, data$Wind, use="[Link]")
mean(data$Ozone, [Link]=T)

##replace the missing data in column

data$Ozone[[Link](data$Ozone)] <- mean(data$Ozone, [Link] = TRUE)
View(data)
data$Ozone <- ifelse([Link](data$Ozone),
mean(data$Ozone, [Link] = T), data$Ozone)
##replace the missing whole data set
data[] <- lapply(data, function(x) {
if ([Link](x)) {
x[[Link](x)] <- mean(x, [Link]= T)
}
return(x)
})