ECO 391 Exam #2 – uky | Actual study
set | Questions and Answers
what does the standard error of the estimate measure? - ANSW-the variability of the observed y-values
around the predicted y-values
what does the R^2 of a multiple regression of y on x1 and x2 measure? - ANSW-the percent variability of
y that is explained by the variability in x1 and x2
What is violation #1 of OLS? - ANSW-model suffers from multicollinearity
What is an example of multicollinearity? - ANSW-sales & sales tax from companies in the same city
What is violation #2 of OLS? - ANSW-heteroskedasticity
What is heteroskedasticity? - ANSW-the circumstance in which the variability of a variable is unequal
across the range of values of a second variable that predicts it
What is an example of heteroskedasticity? - ANSW-predicting sales by square footage of convenience
stores
What is violation #3 of OLS? - ANSW-serial correlation or autocorrelation
What is an example of serial or autocorrelation? - ANSW-quarterly sales @ a restaurant exhibiting
seasonality
What is violation #4 of OLS? - ANSW-Endogeneity
What is endogeneity? - ANSW-when an explanatory variable is correlated w/ the error term, it is called
endogeous and the model is said to suffer from endogeneity
What is an example of endogeneity? - ANSW-omitted variable, reverse causality
What are the 7 assumptions of OLS? - ANSW-1) that the model y=b0+b1x1+...+bkxk is linear in the b
parameters with additive error e
2) conditional on x1, x2...xk the error term has an expected value of zero
3) there's no exact linear relationship among the x values; there's no perfect multicollinearity
4) variance for error term e is the same for all x values (this is called homoskedasticity)
5) error term e is uncorrelated across observations, conditional on explanatory variables (no serial or
autocorrelation)
6) error term e is normally distributed
7) error term e is uncorrelated w/ explanatory variables
What is correlation coefficient? - ANSW-the linear dependence between 2 sets of data
What is the range of the correlation coefficient? - ANSW--1 to 1
As the absolute value of the correlation coefficient approaches 1, what happens to the linear
relationship? - ANSW-it strengthens
when is multicollinearity suspected? - ANSW-when there is a high R^2 coupled with insignificant
explanatory variables
What is a good solution to multicollinearity? - ANSW-drop one of the collinear variables
If the variance of the error term is not the same for all observations, we _____________. - ANSW-cannot
conduct tests of significance
set | Questions and Answers
what does the standard error of the estimate measure? - ANSW-the variability of the observed y-values
around the predicted y-values
what does the R^2 of a multiple regression of y on x1 and x2 measure? - ANSW-the percent variability of
y that is explained by the variability in x1 and x2
What is violation #1 of OLS? - ANSW-model suffers from multicollinearity
What is an example of multicollinearity? - ANSW-sales & sales tax from companies in the same city
What is violation #2 of OLS? - ANSW-heteroskedasticity
What is heteroskedasticity? - ANSW-the circumstance in which the variability of a variable is unequal
across the range of values of a second variable that predicts it
What is an example of heteroskedasticity? - ANSW-predicting sales by square footage of convenience
stores
What is violation #3 of OLS? - ANSW-serial correlation or autocorrelation
What is an example of serial or autocorrelation? - ANSW-quarterly sales @ a restaurant exhibiting
seasonality
What is violation #4 of OLS? - ANSW-Endogeneity
What is endogeneity? - ANSW-when an explanatory variable is correlated w/ the error term, it is called
endogeous and the model is said to suffer from endogeneity
What is an example of endogeneity? - ANSW-omitted variable, reverse causality
What are the 7 assumptions of OLS? - ANSW-1) that the model y=b0+b1x1+...+bkxk is linear in the b
parameters with additive error e
2) conditional on x1, x2...xk the error term has an expected value of zero
3) there's no exact linear relationship among the x values; there's no perfect multicollinearity
4) variance for error term e is the same for all x values (this is called homoskedasticity)
5) error term e is uncorrelated across observations, conditional on explanatory variables (no serial or
autocorrelation)
6) error term e is normally distributed
7) error term e is uncorrelated w/ explanatory variables
What is correlation coefficient? - ANSW-the linear dependence between 2 sets of data
What is the range of the correlation coefficient? - ANSW--1 to 1
As the absolute value of the correlation coefficient approaches 1, what happens to the linear
relationship? - ANSW-it strengthens
when is multicollinearity suspected? - ANSW-when there is a high R^2 coupled with insignificant
explanatory variables
What is a good solution to multicollinearity? - ANSW-drop one of the collinear variables
If the variance of the error term is not the same for all observations, we _____________. - ANSW-cannot
conduct tests of significance
