IFN580 Week 5: Regression Modelling (11%)

Question 1

Logistic regression is a ________ regression technique that is used to model data
having a _____ outcome.

Answer 1

linear, binary

Question 2

Simple regression assumes a __________ relationship between the input attribute(s)
and the output attribute.

Answer 2

linear

Question 3

What is the equation for simple linear regression?

Answer 3

y = b0 + b1x

where b0 is the y-intercept, b1 is the slope and x is the predictor value

Question 4

What type of data are regression models used on?

Answer 4

Linear regression is used on continuous numerical variables, Logistic is used on binary classification

Question 5

Nonlinear regression fits what type of line?

Answer 5

Fits a non-linear function

Question 6

What is R2?

Answer 6

Measures how much total “noise” (variance) is reduced

Question 7

What is Root Mean Squared Error (RMSE)

Answer 7

Measures how much deviation occurs at each point

Lower is better

Question 8

What does it mean if RMSE is high but R² is also high?

Answer 8

The model explains the trend well, but absolute prediction errors are large

Question 9

What is polynomial regression?

Answer 9

A form of regression where the relationship between the independent and dependent variable is modeled as an 𝑛th degree polynomial

Question 10

What’s the formula for residual

Answer 10

Residual = y^ - yi

where y^ = predicted value
yi = observed value
​

Question 11

When do we use Logistic regression?

Answer 11

When the data is binary
when data is always between 0 & 1
when the predicted values are the probability

Question 12

What are the assumptions for Linear Regression?

Answer 12

1. Linearity = there’s a linear relationship b/w the two variables
2. Independence = Residuals are independent
3. Homoscedasticity = residuals have constant variance @ every level of x
4. Normality = residuals follow a normal distribution, centered at 0

Question 13

What are the assumptions for Linear Regression?

Answer 13

1. Curvilinear
2. Independence = Residuals are independent
3. Homoscedasticity = residuals have constant variance @ every level of x
4. Normality = residuals follow a normal distribution, centered at 0

Question 14

Which 𝑅2 value indicates that the line perfectly fits the model?

Answer 14

1

0 would mean that there is no linear relationship

Question 15

The objective of a support vector machine is to find an optimal ________ that best
separates classes from a dataset

Answer 15

hyperplane