Session 9

Question 1

What can you infer from this regression analysis?

Accepted Answer

Age and weekly hours on Internet are dependent (associated)

Question 2

The mean of this normal distribution is 6.29. What is the standard deviation?

Accepted Answer

5.738

Question 3

What is the 95% confidence interval of the population mean weekly hours on Internet for 30 years old people?

Accepted Answer

6.29 plus/minus 2 * (5.738/square root of 200)

Question 4

What is the predicted interval of the weekly hours on Internet considering one random individual 30 years old?

Accepted Answer

6.29 plus/minus 2 * 5.738

Question 5

How much does the value of n° of weekly hours on the Internet change if I increase age by 1?

Accepted Answer

Decrease by 0.168

Question 6

Considering the population of 30 years old people, the weekly number of hours spent on Internet follows a...

Accepted Answer

Normal distribution

Question 7

It results an individual 30 years old spends 6.29 hours on Internet a week.Interpretation considering the population:

Accepted Answer

People 30 years old spend on average (mean) 6.29 hours a week on Internet

Question 8

Why do we look at the histogram of standardized residual?

Accepted Answer

To check that conditional distribution of y at each x value is normal, To check for outliers

Question 9

What is the null hypothesis of the significance test to which the regression analysis refers?

Accepted Answer

The population slope is 0

Question 10

An R-squared equal to 1 means that:

Accepted Answer

Residual sum of squares is equal to 0

Question 11

An R-squared equal to 0 means that:

Accepted Answer

Regression is not useful in predicting the response variable

Question 12

How much do we reduce the prediction error by using regression to predict the weekly hours,instead of its average value?

Accepted Answer

13.34%

Questions 12