Module 4

Question 1

Data scaling improves the accuracy of polynomial prediction especially for higher degrees.

True or False

Answer 1

False

Question 2

What is the meaning of intercept in linear regression?

Answers:
a.
The gradient of the regression line.
b.
The value of the outcome when all of the predictors are 0.

c.
The value of the predictor variable when the outcome is zero.

d.
The relationship between a predictor and the outcome variable.

Answer 2

b.
The value of the outcome when all of the predictors are 0

Question 3

An indication of no linear relationship between two variables would be:

a.
a coefficient of determination equal to 1

b.
a coefficient of determination equal to -1

Correctc.
a coefficient of correlation of 0

d.
a coefficient of correlation equal to -1

e.
both a and b

Answer 3

c.
a coefficient of correlation of 0

Question 4

Point out the wrong statement:

a.
Regression through the origin yields an equivalent slope if you center the data first

b.
Normalizing variables results in the slope being the correlation

c.
Least squares is not an estimation too

Answer 4

c.
Least squares is not an estimation tool

Question 5

which of the following functions are examples of linear models?

f(x) = 2 log x + 3 exp x
g(x) = 2x^2 + 3x
h(x) = 2x + 3

a.
all three f, g and h

b.
h only

c.
g and h only

d.
none of the above

Answer 5

a.
all three f, g and h

Question 6

Which correlation coefficient
ρ
indicates the strongest inverse relationship between two variables:

Answer 6

p = -1

Question 7

The coefficient of determination in linear regression is 0.80. This means:

a.
80% of the variation in x can be explained by regression line

b.
80% of the x values are equal

c.
80% of the y values are positive

d.
80% of the variation in y can be explained by the regression line

Answer 7

d.
80% of the variation in y can be explained by the regression line

Question 8

Which of the following two linear regressions provides a better prediction for points A = (3, 12) and B = (7, 18)?

f(x) = 2x + 5
g(x) = x + 10

a.
function g()

b.
function f()

c.
both have the same predictive accuracy for A and B

Answer 8

c.
both have the same predictive accuracy for A and B

Question 9

Consider fitting a dataset with polynomials of higher and higher degrees. What is the most likely scenario for MAE (mean absolute value) of residuals?

Answers:
a.
increases

b.
decreases

c.
stays the same

Answer 9

b.
decreases

Question 10

The SSE (sum of squared errors) is the same as the sum of squared residuals. True or False

Answer 10

True