dmv-aol.Rmd

---
title: "AOL"
author: "Kimberly Kayla Dewi"
date: "2023-05-08"
output: html_document
---

```{r}
library(readr)
library(rlang)
library(dplyr)
library(visdat)
library(readxl)
library(mlbench)
library(pROC)
library(MLmetrics)
library(rpart)
library(rpart.plot)
#library(party)
library(randomForest)
#library(caret)
```
memanggil library

#GDP by Country dataset
library xlsx
```{r}
library(readxl)
```
memanggil library

```{r}
library(visdat)
```
memanggil library

```{r}
gdp_data <- read_xlsx('gdp-in-us.xlsx')
```
membaca file xlsx

```{r}
View(gdp_data)
```
melihat file dataset

```{r}
library(dplyr)
```
install library

```{r}
gdp_reduced <- gdp_data %>%  filter(!row_number() %in% c(1, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61))
```
menghapus tahun 1961-1992, 2004-2014, dan 2021

```{r}
View(gdp_reduced)
```
melihat data gdp_reduced

##General Characteristics
```{r}
head(gdp_reduced)
```
display head data

```{r}
tail(gdp_reduced)
```
display tail data

```{r}
View(gdp_reduced)
```
melihat data

```{r}
sapply(gdp_reduced, class)
```
display tipe data

```{r}
str(gdp_reduced)
```
display karakteristik data

```{r}
attributes(gdp_reduced)
```
attribut dr data

```{r}
unique(gdp_reduced)
```
list data yg unik pd tiap kolom

```{r}
View(gdp_reduced)
```
melihat data

```{r}
colSums(is.na(gdp_reduced))
```
mengetahui jumlah data yg kosong di tiap kolom

```{r}
dim(gdp_reduced)
```
mengetahui jumlah baris dan kolom dari data

```{r}
vis_miss(gdp_reduced)
```
persentase data sblm dibersihkan
tidak ada data yg kosong

```{r}
BasicSummary <- function(df, dgts = 3)
{
  m <- ncol(df)
  varNames <- colnames(df)
  varType <- vector("character", m)
  topLevel <- vector("character", m)
  topCount <- vector("numeric", m)
  missCount <- vector("numeric", m)
  levels <- vector("numeric", m)
  
  for (i in 1:m)
  {
    x <- df[,i]
    varType[i] <- class(x)
    xtab <- table(x, useNA = "ifany")
    levels[i] <- length(xtab)
    nums <- as.numeric(xtab)
    maxnum <- max(nums)
    topCount[i] <- maxnum
    maxIndex <- which.max(nums)
    lvls <- names(xtab)
    topLevel[i] <- lvls[maxIndex]
    missIndex <- which((is.na(x)) | (x == "") | (x == " "))
    missCount[i] <- length(missIndex)
  }
  n <- nrow(df)
  topFrac <- round(topCount/n, digits = dgts)
  missFrac <- round(missCount/n, digits = dgts)
  summaryFrame <- data.frame(variable = varNames, type = varType,
  levels = levels, topLevel = topLevel,
  topCount = topCount, topFrac = topFrac,
  missFreq = missCount, missFrac = missFrac)
  return(summaryFrame)
  }
```
basic summary code

```{r}
BasicSummary(gdp_reduced)
```
basic summary dr data

##Descriptive Statistics
```{r}
summary(gdp_reduced)
```
summary dr data

```{r}
var(gdp_reduced)
```
varians data

```{r}
apply(gdp_reduced[1:4], 2, sd, na.rm = T)
```
standar deviasi data

##Data Anomalies/Outliers
```{r}
boxplot(year ~ gdp, data = gdp_reduced)
```
tidak terdapat outlier pada data

```{r}
table(gdp_reduced)
```
table yang berisi list outlier pada dataset

##Exploratory Visualization
```{r}
hist(gdp_reduced$year)
```
histogram frekuensi tahun yang terpilih untuk diteliti

```{r}
hist(gdp_reduced$gdp)
```
histogram  frekuensi gdp

```{r}
hist(gdp_reduced$gdp_per_capita)
```
histogram gdp_per_capita beserta frekuensinya

```{r}
hist(gdp_reduced$growth)
```
histogram growth beserta frekuensinya

##Explanatory Visualization
```{r}
plot(gdp_reduced)
```
scatterplot gdp_reduced

##Linear Regression
```{r}
par(mfrow = c(1,1))
x <- gdp_reduced$gdp
y <- gdp_reduced$year
plot(x, y, xlab = "gdp", ylab = "year")
index <- which(y > 0)
points(x[-index], y[-index], pch = 16) 
points(x[-index], y[-index], pch = 18, cex = 2)
olsmodel <- lm(y ~ x)
abline(olsmodel, lty = 3)
```
linear regression pada gdp per capita dan growth

##Predictive Model
```{r}
keepVars <- c("year", "gdp", "gdp_per_capita", "growth")
gdpSub <- gdp_reduced[, keepVars]
```
untuk menyimpan nilai pada variabel

###Training set and Validation set
```{r}
set.seed(123)
n <- nrow(gdpSub)
train <- sample(n, round(0.7 * n))
gdpTrain <- gdpSub[train,]
gdpValidation <- gdpSub[-train,]
```
membuat training set dan validation set

```{r}
View(gdpTrain)
View(gdpValidation)
```
melihat data training dan validation set

```{r}
logisticFull <- glm(gdp ~., family = "poisson", data = gdpTrain) #poisson/binomial/gaussian
summary(logisticFull)
```
summary logisticFull untuk gdp

```{r}
exp(coef(logisticFull))
```
exponentiating the logisticFull outcome

```{r}
logisticRef <- glm(gdp ~ growth, family = "poisson", data = gdpTrain)
summary(logisticRef)
```
summary logisticRef untuk gdp

```{r}
exp(coef(logisticRef))
```
exponentiating the logisticRef outcome

```{r}
gdp_pHatFullv <- predict(logisticFull, newdata = gdpValidation, type = "response")
gdp_pHatRefv <- predict(logisticRef, newdata = gdpValidation, type = "response")
```
validation model

```{r}
gdp_pHatRefv
```
hasil gdp_pHatRefv

```{r}
gdpValidation$gdp
```
hasil gdpValidation$gdp

###Accuracy Model
```{r}
threshold1 <- table(gdpValidation$gdp, gdp_pHatFullv > 0.5)
threshold2 <- table(gdpValidation$gdp, gdp_pHatRefv > 0.5)
accuracymodel1 <- round(sum(diag(threshold1)) / sum(threshold1), 2)
accuracymodel2 <- round(sum(diag(threshold2)) / sum(threshold2), 2)
sprintf("Accuracy model 1 is %s", accuracymodel1)
sprintf("Accuracy model 2 is %s", accuracymodel2)
```
accuracy model 1 and 2 sama

###Predictive Model
```{r}
library(randomForest)
```
install library

```{r}
PreModel1 <- predict(gdpModel1, newdata = gdpValidation, type = "class")
PreModel2 <- predict(gdpModel2, newdata = gdpValidation, type = "class")
```
membuat variabel baru

```{r}
library(rpart)
library(rpart.plot)
#library(party)
```
install library

```{r}
gdpModel1 <- rpart(gdp ~., data = gdpTrain, method = "class")
gdpModel2 <- rpart(gdp ~ gdp_per_capita, data = gdpTrain, method = "class")
```
membuat variabel baru

###Confusion Matrix for the 1st Model
```{r}
cm <- table(PreModel1, gdpValidation$gdp)
cm
```
membuat variabel confusion matrix

####High
```{r}
accuracy <- sum(cm[1], cm[5:6], cm[8:9]) / sum(cm[1:9])
precision <- cm[1] / sum(cm[1:3])
sensitiv <- cm[1] / sum(cm[1], cm[4], cm[7])
fscore <- (2 * (sensitiv * precision))/(sensitiv + precision)
specificity <- sum(cm[5:6], cm[8:9]) / sum(cm[5:6], cm[8:9], cm[2:3])
sprintf("Accuracy is %s", round(accuracy,3))
sprintf("Precision is %s", round(precision,3))
sprintf("sensitivity is %s", round(sensitiv,3))
sprintf("specificity is %s", round(specificity,3))
```
high predictive model

```{r}
gdpModel1
```
model 1 gdp

####Medium
```{r}
accuracy <- sum(cm[9], cm[1], cm[2], cm[4], cm[5]) / sum(cm[1:9])
precision <- cm[9] / sum(cm[7:9])
sensitiv <- cm[9] / sum(cm[9], cm[3], cm[6])
fscore <- (2 * (sensitiv * precision))/(sensitiv + precision)
specificity <- sum(cm[1], cm[2], cm[4], cm[5]) / sum(cm[1], cm[2], cm[4], cm[5], cm[7], cm[8])
sprintf("Accuracy is %s", round(accuracy,3))
sprintf("Precision is %s", round(precision,3))
sprintf("sensitivity is %s", round(sensitiv,3))
sprintf("specificity is %s", round(specificity,3))
```
medium predictive model

####Low
```{r}
accuracy <- sum(cm[5], cm[1], cm[3], cm[7], cm[9]) / sum(cm[1:9])
precision <- cm[5] / sum(cm[4:6])
sensitiv <- cm[5] / sum(cm[2], cm[5], cm[8])
fscore <- (2 * (sensitiv * precision))/(sensitiv + precision)
specificity <- sum(cm[1], cm[3], cm[7], cm[9]) / sum(cm[1], cm[3], cm[7], cm[9], cm[4], cm[6])
sprintf("Accuracy is %s", round(accuracy,3))
sprintf("Precision is %s", round(precision,3))
sprintf("sensitivity is %s", round(sensitiv,3))
sprintf("specificity is %s", round(specificity,3))
```
low predictive model

###Confusion Matrix for the 2nd Model

####High
```{r}
accuracy <- sum(cm[1], cm[5:6], cm[8:9]) / sum(cm[1:9])
precision <- cm[1] / sum(cm[1:3])
sensitiv <- cm[1] / sum(cm[1], cm[4], cm[7])
fscore <- (2 * (sensitiv * precision))/(sensitiv + precision)
specificity <- sum(cm[5:6], cm[8:9]) / sum(cm[5:6], cm[8:9], cm[2:3])
sprintf("Accuracy is %s", round(accuracy,3))
sprintf("Precision is %s", round(precision,3))
sprintf("sensitivity is %s", round(sensitiv,3))
sprintf("specificity is %s", round(specificity,3))
```
high predictive model

####Medium
```{r}
accuracy <- sum(cm[9], cm[1], cm[2], cm[4], cm[5]) / sum(cm[1:9])
precision <- cm[9] / sum(cm[7:9])
sensitiv <- cm[9] / sum(cm[9], cm[3], cm[6])
fscore <- (2 * (sensitiv * precision))/(sensitiv + precision)
specificity <- sum(cm[1], cm[2], cm[4], cm[5]) / sum(cm[1], cm[2], cm[4], cm[5], cm[7], cm[8])
sprintf("Accuracy is %s", round(accuracy,3))
sprintf("Precision is %s", round(precision,3))
sprintf("sensitivity is %s", round(sensitiv,3))
sprintf("specificity is %s", round(specificity,3))
```
medium predictive model

```{r}
gdpModel2
```
model 2 for gdp

#GDP Year
library xlsx
```{r}
library(readxl)
```
memanggil library

```{r}
library(visdat)
```
memanggil library

```{r}
GDP_data <- read_xlsx('GDP.xlsx')
```
membaca file xlsx

```{r}
View(GDP_data)
```
melihat file dataset

```{r}
library(dplyr)
```
install library

```{r}
GDP_data <- GDP_data[,c("United States")]
GDP_reduced <- GDP_data[669:679,]
Year = c(1913, 1914, 1915, 1916, 1917, 1918, 1919, 1920, 1921, 1922, 1923)
GDP_reduced$Year <- Year
GDP_reduced
```
mengambil tahun 1913-1923

```{r}
View(GDP_reduced)
```
melihat data GDP_reduced

##General Characteristics
```{r}
head(GDP_reduced)
```
display head data

```{r}
tail(GDP_reduced)
```
display tail data

```{r}
sapply(GDP_reduced, class)
```
display tipe data

```{r}
str(GDP_reduced)
```
display karakteristik data

```{r}
attributes(GDP_reduced)
```
attribut dr data

```{r}
unique(GDP_reduced)
```
list data yg unik pd tiap kolom

```{r}
View(GDP_reduced)
```
view data

```{r}
colSums(is.na(GDP_reduced))
```
mengetahui jumlah data yg kosong di tiap kolom

```{r}
dim(GDP_reduced)
```
mengetahui jumlah baris dan kolom dari data

```{r}
vis_miss(GDP_reduced)
```
persentase data sblm dibersihkan
tidak ada data yg kosong

```{r}
BasicSummary <- function(df, dgts = 3)
{
  m <- ncol(df)
  varNames <- colnames(df)
  varType <- vector("character", m)
  topLevel <- vector("character", m)
  topCount <- vector("numeric", m)
  missCount <- vector("numeric", m)
  levels <- vector("numeric", m)
  
  for (i in 1:m)
  {
    x <- df[,i]
    varType[i] <- class(x)
    xtab <- table(x, useNA = "ifany")
    levels[i] <- length(xtab)
    nums <- as.numeric(xtab)
    maxnum <- max(nums)
    topCount[i] <- maxnum
    maxIndex <- which.max(nums)
    lvls <- names(xtab)
    topLevel[i] <- lvls[maxIndex]
    missIndex <- which((is.na(x)) | (x == "") | (x == " "))
    missCount[i] <- length(missIndex)
  }
  n <- nrow(df)
  topFrac <- round(topCount/n, digits = dgts)
  missFrac <- round(missCount/n, digits = dgts)
  summaryFrame <- data.frame(variable = varNames, type = varType,
  levels = levels, topLevel = topLevel,
  topCount = topCount, topFrac = topFrac,
  missFreq = missCount, missFrac = missFrac)
  return(summaryFrame)
  }
```
basic summary code

```{r}
BasicSummary(GDP_reduced)
```
basic summary dr data

##Descriptive Statistics
```{r}
summary(GDP_reduced)
```
summary dr data

```{r}
var(GDP_reduced)
```
varians data

```{r}
apply(GDP_reduced[1:2], 2, sd, na.rm = T)
```
standar deviasi data

##Data Anomalies/Outliers
```{r}
boxplot(GDP_reduced)
```
tidak terdapat outlier pada data

```{r}
table(GDP_reduced)
```
table yang berisi list outlier pada dataset

##Exploratory Visualization
```{r}
hist(GDP_reduced$`United States`)
```
histogram frekuensi tahun yang terpilih untuk diteliti

##Explanatory Visualization
```{r}
plot(GDP_reduced)
```
scatterplot gdp_reduced

##Linear Regression
```{r}
par(mfrow = c(1,1))
x <- GDP_reduced$`United States`
y <- GDP_reduced$Year
plot(x, y, xlab = "gdp", ylab = "Year")
index <- which(y > 0)
points(x[-index], y[-index], pch = 16) 
points(x[-index], y[-index], pch = 18, cex = 2)
olsmodel <- lm(y ~ x)
abline(olsmodel, lty = 3)
```
linear regression pada gdp per capita dan growth