## Exercises

Use the multivariate dataset named Auto MPG from the UCI Machine Learning repository (reference 3 in the

*References*section of this chapter). The dataset can be downloaded from the website at https://archive.ics.uci.edu/ml/datasets/Auto+MPG. The dataset describes automobile fuel consumption in**miles per gallon**(**mpg**) for cars running in American cities. From the folder containing the datasets, download two files:`auto-mpg.data`

and`auto-mpg.names`

. The`auto-mpg.data`

file contains the data and it is in space-separated format. The`auto-mpg.names`

file has several details about the dataset, including variable names for each column. Build a regression model for the fuel efficiency, as a function displacement (*disp*), horse power (*hp*), weight (*wt*), and acceleration (*accel*), using both OLS and Bayesian GLM. Predict the values for mpg in the test dataset using both the OLS model and Bayesian GLM model (using the`bayesglm`

function). Find the**Root Mean Square Error**(**RMSE**) values for OLS and...