Regression Classification Analysis Quiz

Regression analysis is primarily used for:

• Classification of data

• Predicting a continuous dependent variable

• Clustering observations

• Dimension reduction

In a simple linear regression model, the relationship is assumed to be:

• Quadratic

• Linear between dependent and independent variable

• Exponential

• Logistic

Which of the following is an assumption of linear regression?

• Independence of errors

• Homoscedasticity (constant variance of errors)

• Linearity between dependent & independent variables

• All of the above

Multiple Linear Regression involves:

• One independent and one dependent variable

• More than one dependent variable

• Multiple independent variables and one dependent variable

• No variables

Which regression is best suited for categorical dependent variables?

• Linear Regression

• Logistic Regression

• Ridge Regression

• Polynomial Regression

Polynomial Regression is used when:

• Data is normally distributed

• Relationship is nonlinear but can be modeled as polynomial terms

• Only one variable exists

• Multicollinearity is absent

Gradient Descent is used in regression for:

• Scaling features

• Minimizing the cost function

• Increasing correlation

• Normalizing data

The learning rate in Gradient Descent controls:

• Speed of convergence

• Number of variables

• Size of dataset

• Degree of polynomial

If the learning rate is too high in gradient descent, the algorithm may:

• Converge faster

• Overshoot and fail to converge

• Always find the global minimum

• Stop immediately

Correlation measures:

• Causal relationship

• Strength and direction of linear relationship

• Nonlinear dependencies

• Cost minimization