Practical Data Science using Python - Linear Regression Data Preparation and Analysis 3

The video tutorial covers visual analysis of car datasets, focusing on creating a derived dataset called cars ALR. It explains the use of pair plots to identify linear relationships between features and discusses the importance of identifying and handling multicollinearity in linear regression models. The tutorial also covers feature engineering by converting categorical variables into dummy variables, ensuring they are suitable for regression analysis.

What is the primary purpose of using a pair plot in data analysis?

• To identify missing values in the dataset

• To visualize relationships between pairs of features

• To calculate the mean of all features

• To sort data in ascending order

Why is it important to identify highly correlated features in a dataset?

• To increase the number of features

• To ensure data is normally distributed

• To improve data visualization

• To avoid multicollinearity in regression models

What does a high degree of correlation between two features indicate?

• They are independent of each other

• They have a strong linear relationship

• They are categorical variables

• They have no relationship

What is the purpose of using a heat map in data analysis?

• To identify outliers in the dataset

• To visualize the correlation between multiple features

• To perform data normalization

• To display the distribution of a single variable

What is the Variance Inflation Factor (VIF) used for?

• To measure the variance of a single feature

• To identify multicollinearity among features

• To calculate the mean of all features

• To visualize data distribution

Why are dummy variables created from categorical variables?

• To eliminate missing values

• To increase the complexity of the model

• To reduce the number of features

• To convert categorical data into a numeric format

What is a key consideration when creating dummy variables from ordinal categorical variables?

• Minimizing the number of dummy variables

• Maximizing the number of dummy variables

• Avoiding the assumption of inherent order

• Ensuring they have an inherent order

How many dummy variables are created for a categorical feature with five possible values?

• Two

• Three

• Four

• Five

What is the main goal of Principal Component Analysis (PCA)?

• To perform data normalization

• To reduce the number of dimensions while retaining information

• To increase the number of features

• To eliminate all categorical variables

What happens to the number of features after converting categorical variables to dummy variables?

• The number of features remains the same

• The number of features becomes zero

• The number of features decreases

• The number of features increases