Pre-processing the housing data
The dataset was selected from www.kaggle.com and the title of the project is House Prices: Advanced Regression Techniques. The main files we will be using are test.csv and train.csv, and the files are available in the companion bundle package. A description of the variables can be found in the data_description.txt file. Further details, of course, can be obtained at https://www.kaggle.com/c/house-prices-advanced-regression-techniques/. The train dataset contains 1460 observations, while the test dataset contains 1459 observations. The price of the property is known only in the train dataset and are not available for those in the test dataset. We will use the train dataset for model development only. The datasets are first loaded into an R session and a beginning inspection is done using the read.csv, dim, names, and str functions:
> housing_train <- read.csv("../Data/Housing/train.csv",
+ row.names = 1,na.strings ...
