Imputing missing data

The previous recipe showed us how to detect missing values within the dataset. Though the data with missing values is rather incomplete, we can still adapt a heuristic approach to complete our dataset. Here, we introduce some techniques one can employ to impute missing values.

Getting ready

Refer to the Converting data types recipe and convert each attribute of imported data into the proper data type. Also, rename the columns of the employees and salaries datasets by following the steps in the Renaming the data variable recipe.

How to do it…

Perform the following steps to impute missing values:

1. First, we subset user data with emp_no equal to 10001:

   > test.emp <- salaries[salaries$emp_no == 10001,]
   

2. Then, we purposely assign salary as the missing value of row 8:

   > test.emp[8,c("salary")]
   [1] 75286
   > test.emp[8,c("salary")] = NA
   

3. For the first imputing method, we can remove records with missing values using the na.omit function:

   > na.omit(test.emp)
   

4. On the other...