Which of the following is not one of the assumptions of regression?
a. There is a population regression line that joins the SDs of all possible distributions of results.
b. The response variable is normally distributed.
c. The errors are probabilistically
d. The standard deviation of the response variable increases as the explanatory variables increase.
A scatterplot that exhibits a "fan" shape (the variation of Y increases as X increases) is an example of:
a. heteroscedasticity
b. multicollinearity
c. autocorrelation
d. homoscedasticity
a. heteroscedasticity
The appropriate hypothesis test for a regression coefficient is:
a. H0: ? � 0, Ha: ? = 0
b. H0: ? = 0, Ha: ? � 0
c. H0: ? = 1, Ha: ? � 1
d. None of these choices are correct.
b. H0: ? = 0, Ha: ? � 0
A researcher can check whether the errors are normally distributed by using:
a. a frequency distribution or the value of the regression coefficient
b. the Durbin-Watson statistic
c. a t-test or an F-test
d. a histogram or a Q-Q plot
d. a histogram or a Q-Q plot
Suppose you forecast the values of all of the independent variables and insert them into a multiple regression equation and obtain a point prediction for the dependent variable. You could then use the standard error of the estimate to obtain an approximat
c. prediction interval
Forward regression:
a. adds and deletes variables until an optimal equation is achieved
b. begins with no explanatory variables in the equation and successively adds one at a time until no remaining variables make a significant contribution
c. begins with
b. begins with no explanatory variables in the equation and successively adds one at a time until no remaining variables make a significant contribution
If you can determine that the outlier is not really a member of the relevant population, then it is appropriate and probably best to:
a. delete it
b. average it
c. leave it
d. reduce it
a. delete it
When the error variance is nonconstant, it is common to see the variation increases as the explanatory variable increases (you will see a "fan shape" in the scatterplot). There are two ways you can deal with this phenomenon. These are:
a. the partial F an
d. the weighted least squares and a logarithmic transformation
The t-value for testing H0: ?1 = 0 is calculated using which of the following equations?
a. n - k - 1
b. ?i/Si
c. bi/Sbi
d. ?(Xi/Yi)
c. bi/Sbi
Suppose you run a regression of a person's height on his/her right and left foot sizes, and you suspect that there may be multicollinearity between the foot sizes. What types of problems might you see if your suspicions are true?
a. "wrong" values for the
d. All of these choices are correct.
The appropriate hypothesis test for an ANOVA test is:
a. H0: at least one ? = 0, Ha: all ? � 0
b. H0: all ? � 0, Ha: at least one ? = 0
c. H0: all ? = 0, Ha: at least one ? � 0
d. H0: at least one ? � 0, Ha: all ? = 0
c. H0: all ? = 0, Ha: at least one ? � 0
Which approach can be used to test for autocorrelation?
a. correlation coefficient
b. Durbin-Watson statistic
c. F-test or t-test
d. regression coefficient
b. Durbin-Watson statistic
An error term represents the vertical distance from any point to the:
a. population regression line
b. mean value of the X's
c. estimated regression line
d. value of the Y's
a. population regression line
There is evidence that the regression equation provides little explanatory power when the F-ratio:
a. is large
b. is the constant
c. equals the regression coefficient
d. is small
d. is small
Which definition best describes parsimony?
a. explaining the most with the least
b. being able to explain all of the change in the response variable
c. being able to predict the value of the response variable far into the future
d. explaining the least wi
a. explaining the most with the least