Question

1. In least squares regression, which of the following is not a required assumption about the error term ε? 

a. The expected value of the error term is one. 

b. The variance of the error term is the same for all values of x. 

c. The values of the error term are independent. 

d. The error term is normally distributed. 

2. Larger values of R² imply that the observations are more closely grouped about the 

a. Average value of the independent variables 

b. Average value of the dependent variable 

c. Least squares line 

d. Origin

3. A correlation between age and health of a person found to be -1.09. On the basis of this you would tell the doctors that: 

a. The age is good predictor of health 

b. The age is poor predictor of health 

c. None of these 

4. In a regression analysis if R²=1, then 

a. SSg must also be equal to one 

b. SSg must be equal to zero 

c. SSg can be any positive value 

d. SSg must be negative 

5. The least squares regression line minimizes the sum of the 

a. Differences between actual and predicted Y values 

b. Absolute deviations between actual and predicted Y values 

c. Absolute deviations between actual and predicted X values 

d. Squared differences between actual and predicted Y values 

e. Squared differences between actual and predicted X values

Answer 1

Que.1

Option a is the answer.

Because in regression we assume that errors are independent normally distributed with mean zero and constant variance σ²

Que.2

Option c is correct.

The coefficient of determination (R2) is a measure of the proportion of variance of a predicted outcome. A value of 1 means every point on the regression line fits the data.

Que.3

Option a is correct.

Value of correlation coefficient is 1 means it is prrfect correlation.

Que.4

Option b is correct.R² = SSR/SST

Where SST = SSR +SSE, if SSE = 0 then SST=SSR and R²=1

Que.5

Option d is correct.