desc_stats.Rmd

---
title: "Descriptive statistics"
output:
  pdf_document: default
  html_notebook: default
---

```{r}
RM <- read.csv("finalData.csv",header=TRUE,sep = ";")
head(RM)
str(RM)
```

```{r}
library(lattice)
par(mfrow = c(2,2))
df <- unique(RM[c(6,8:17)])
df$DefaultedLoans <- as.numeric(df$DefaultedLoans)
head(df)
```
```{r}
var_list<- names(df)[2:11]
models <- lapply(var_list, 
          function(x) 
            {
             mod = lm(substitute(DefaultedLoans ~ i, list(i = as.name(x))), data = df)
            })
par(mfrow = c(2,2))
invisible(lapply(models,plot))
```

```{r}
library(ggplot2)
plots <- function(i){
                     ggplot(df, aes(x = df[i])) + 
                     geom_histogram(aes(y = ..density..),colour = "black", 
                                    fill = "yellow",binwidth = 0.02) +
                     geom_density(alpha = 0.2, fill = "blue",linetype = "dashed") + 
                     geom_vline(aes(xintercept=mean(df[i], na.rm = TRUE)),
                                color = "black",linetype = "dashed", size = 1) +
                     xlab(names(df)[i]) +
                     ggtitle(paste("Histogram plot of ",names(df)[i]))}
par(mfrow = c(2,2))
lapply(2:11,FUN = plots)
```
```{r}
library(s20x)
pairs(df,col="green",pch = 20)
pairs20x(df)
```
```{r}
library(stargazer)
library(doBy)
stargazer(df, type = "text", median = TRUE, digits = 2)
summaryBy(AddressLatitude ~ DefaultedLoans, data = RM, FUN = c(mean), na.rm =TRUE)
summaryBy(AddressLongitude ~ DefaultedLoans, data = RM, FUN = c(mean), na.rm =TRUE)
summaryBy(CreditRating ~ DefaultedLoans, data = RM, FUN = c(mean), na.rm =TRUE)
summaryBy(CreditRating ~ County, data = RM, FUN = c(mean), na.rm =TRUE)
```
```{r}
reg <- lm(df$DefaultedLoans~df$CreditRating+df$MortgageYears+df$CreditRatingMovement+df$LTV+df$LoanBalance+df$InterestIncome+df$PropertyValue+df$AnnualPYMT+df$AddressLatitude+df$AddressLongitude)
summary(reg)
```