DS_midterm_part1.Rmd

```{r}
dt = read.csv("C:\\Users\\91799\\Downloads\\college.csv")
```


```{r}
View(dt)
```


```{r}
summary(dt)
```


```{r}
dt$Private <- ifelse(dt$Private == "Yes",1,0)
pairs(dt[,2:11])

```


```{r}
par(mfrow=c(4,4), mar=c(2, 2, 1, 0))
for (i in 2:17)
  boxplot(dt$Outstate ~ dt$Private, main=colnames(dt[i])) 

```


```{r}
dt$Elite <- dt$Top10perc > 50
dt$Elite <- ifelse(dt$Elite == "TRUE",1,0)

```


```{r}
summary(dt[, c("Top10perc", "Elite")])
boxplot(Outstate ~ Elite, data=dt)

```


```{r}
par(mfrow=c(4,4))
for (i in 2:17)
  hist(dt[,i], xlab = colnames(dt)[i], breaks = 10)

```

#Exploring which universities have the most and least Top25perc, Most S.F. Ratio and plotiing Graduation rate and Top10perc to get an idea how they co-relate
```{r}
row.names(dt)[which.max(dt$Top25perc)]
row.names(dt)[which.min(dt$Top25perc)]
plot(dt$Top10perc, dt$Grad.Rate)
row.names(dt)[which.max(dt$S.F.Ratio)]
```




```{r}
auto = read.csv("C:\\Users\\91799\\Downloads\\auto.csv")
auto<- na.omit(auto)
dim(auto)
summary(auto)
View(auto)
```


```{r}
sapply(auto,class)
auto$origin <- factor(auto$origin, levels=1:3, labels=c("U.S.", "Europe", "Japan"))
sapply(auto,class)
quant <- sapply(auto, is.numeric)
quant

```


```{r}
sapply(auto[,quant], range)

```


```{r}
sapply(auto[,quant], function(x) round(c(mean(x), sd(x)), 3))
```


```{r}
result<-sapply(auto[-10:-85,quant], function(x) round(c(range(x), mean(x), sd(x)), 3))
result

```
We can see heavier weight corelates with lower mpg, more cylinders less mpg and also over time cars become more effecient

```{r}
pairs(auto[,1:7])

```

#All predictors show some kind of corelation with mpg. It is difficult to generalize but we can see that years, origin and other variables affect mpg
```{r}

with(auto, plot(mpg, year))
with(auto, plot(origin, mpg), ylab= "mpg")

```