[Solved] DataMining Homework1

Homework 1

• (R programming + Data Processing) Consider the College dataset in the package ISLR.
  1. Use the function summary() to produce a numerical summary of the variables in the dataset.
  2. Use pairs() to produce a scatterplot of the continuous variables in the data set.
  3. Create a new qualitative variable called Elite, by binning the variable Top10perc. We are going to divide universities into two groups based on whether or not the proportion of students coming from the two 10% of their high school exceeds 50%. Add this variable to your dataset.
  4. Use the table function to figure out how many Elite schools there are.
  5. Use the table function to figure out how many of the Elite schools are private.
  6. Do elite schools tend to have higher graduation rates?

• (R programming + Data Processing

This exercise uses the Auto dataset in the package ISLR. a) Remove missing values from the data.

1. What variables are numerical (continuous) or factors (categorical)?
2. Report the mean and standard deviation for each continuous variable in the data.
3. Remove the 5^th through 55^th What is the range, mean and standard deviation?
4. In the full Auto dataset, are there any variables you would consider removing, or representing differently? Why?
5. In the full Auto dataset, graphically explore the relationships between the variables in the data set.
6. In the full Auto dataset, consider the variable mpg. You are going to create a new categorical variable for MPG, which has the categories: {low, med, high}. Call this variable my_mpg, and create a new_Auto dataset, which contains all of the Auto variables, and your new variable my_mpg. Save the dataset as an *.RData file and submit it with your assignment.