### (a) Investigate the distribution of the variable - Emissions.NOx.mg.km â€“ using suitable plots. Are there any outlying values? Should you use a transformation of this variable? If so which one would you use? Give reasons for your choices.
```{r, fig.align='center'}
boxplot(my.data$Emissions.NOx.mg.km,
   xlab = "Emissions.NOx.mg.km",
   ylab = "Values", main = "Emissions.NOx.mg.km boxplot")
```

From the boxplot it is evident that there is a potential outlier having a very high value. It is too far from the 3rd quartile of the data. It can be removed just by cutting-off data above 75th percentile.

```{r, fig.align='center'}
quant_range <- quantile(my.data$Emissions.NOx.mg.km)
emission.without.outlier <- my.data$Emissions.NOx.mg.km[my.data$Emissions.NOx.mg.km <= quant_range[4]]
boxplot(emission.without.outlier,
   xlab = "Emissions.NOx.mg.km",
   ylab = "Values", main = "Emissions.NOx.mg.km without outlier")
```

### (b) Form histograms of both the Engine.Capacity and Noise.Level.dB.A separately for the different levels in the variable (Fuel.Type). Provide an interpretation of the distributions in the histograms.

```{r, fig.align='center'}
par(mfrow=c(2,1))
hist(my.data$Engine.Capacity[my.data$Fuel.Type == "Diesel"], xlab = "Engine Capacity (Diesel)", main = "Engine Capacity of Diesel Car")
hist(my.data$Engine.Capacity[my.data$Fuel.Type == "Petrol"], xlab = "Engine Capacity (Petrol)",
main = "Engine Capacity of Petrol Car")

par(mfrow=c(2,1))
hist(my.data$Noise.Level.dB.A[my.data$Fuel.Type == "Diesel"], xlab = "Engine Capacity (Diesel)",
main = "Noise.Level.dB.A of Diesel Car")
hist(my.data$Noise.Level.dB.A[my.data$Fuel.Type == "Petrol"], xlab = "Engine Capacity (Petrol)",
main = "Noise.Level.dB.A of Petrol Car")
```

From the histogram it is observed that the distribution of Engine.Capacity is right skewed irrespective of fuel type. Engine capacity is mainly concentrated at level 1500 and 2000.

Distribution of Noise.Level.dB.A is normally distributed for diesel car # while it is slightly left skewed for petrol cars.

Use R for Q1-Q3 1. Use R to randomly assign 10 experimental units...