BRAINSCAPE1
BUSN 4000 - Final Exam with
complete solutions latest version
residual - CORRECT ANSWER-difference between the actual y value and the predicted
value of y at point x - represents the "error"
s(e) - CORRECT ANSWER-estimate of the standard deviation around the regression
line - standard error of regression
standard error - CORRECT ANSWER-typical sized error of the prediction when using
the regression model
Hypothesis test for a (single) slope coefficient - CORRECT ANSWER-Ho: Bi = 0
Ha: Bi =/= 0
Hypothesis test for a slope coefficient - interpretation - CORRECT ANSWER-Ho: Not
Useful
Ha: Useful
Hypothesis test for a (single) slope coefficient - test - CORRECT ANSWER-T test
test statistic - CORRECT ANSWER-value you can calculate from the sample data that
measures how much evidence your sample data gives you to dispute the assumption in
the null hypothesis
p-value - CORRECT ANSWER-probability that you calculate from your test statistic -
small p-value = reject the null
R2 - CORRECT ANSWER-- what proportion of the total variation in the response
variable has been explained by the model
- the closer to 1 the value of R2 is, the better "fit" of the regression line to the data
BRAINSCAPE1
, BRAINSCAPE1
R2 Interpretation - CORRECT ANSWER-If R2=.8872 you can said the variable explains
approximately 88.72% of the variation in the dependent variable.
Adjusted R2 - CORRECT ANSWER-- penalizes for # of variables used
- better indicator of model fit in multiple regression and can be used to inform decisions
Hypothesis test for a multiple variables - CORRECT ANSWER-Ho: B1=B2=B3=0
Ha: At least one coefficient is not equal to zero
Hypothesis test for a multiple variables - interpretation - CORRECT ANSWER-Ho: None
of the explanatory variables are useful
Ha: At least one of the explanatory variables are useful
Hypothesis test for a multiple variables - test - CORRECT ANSWER-F test
F-Test - CORRECT ANSWER-- global test assessing the overall fit of regression (not
specific variables)
- large F means small P
- Small F means large P
Hypothesis test for a reduced model - CORRECT ANSWER-Ho: Reduced model
variables = 0
Ha: At least one of the reduced model variables is not equal to zero
Hypothesis test for a reduced model - interpretation - CORRECT ANSWER-Ho: chose
the reduced model
Ha: at least one of the variables is contributing to the explanation of ht variation in y and
the full model is chosen as superior to the reduced
Hypothesis test for a reduced model - test - CORRECT ANSWER-partial F-test
Confidence Interval - CORRECT ANSWER-- an interval estimate of the average
predicted value for a number of observations
- "average" for "all"
- narrower
Prediction Interval - CORRECT ANSWER-- interval estimate of a predicted value for a
single observation
- "single" territory
- wider
Factors that influence CI and PI - CORRECT ANSWER-- sample size (larger =
narrower)
- spread of data (smaller error = narrower)
- confidence interval (99% is wider than 90%)
BRAINSCAPE1
BUSN 4000 - Final Exam with
complete solutions latest version
residual - CORRECT ANSWER-difference between the actual y value and the predicted
value of y at point x - represents the "error"
s(e) - CORRECT ANSWER-estimate of the standard deviation around the regression
line - standard error of regression
standard error - CORRECT ANSWER-typical sized error of the prediction when using
the regression model
Hypothesis test for a (single) slope coefficient - CORRECT ANSWER-Ho: Bi = 0
Ha: Bi =/= 0
Hypothesis test for a slope coefficient - interpretation - CORRECT ANSWER-Ho: Not
Useful
Ha: Useful
Hypothesis test for a (single) slope coefficient - test - CORRECT ANSWER-T test
test statistic - CORRECT ANSWER-value you can calculate from the sample data that
measures how much evidence your sample data gives you to dispute the assumption in
the null hypothesis
p-value - CORRECT ANSWER-probability that you calculate from your test statistic -
small p-value = reject the null
R2 - CORRECT ANSWER-- what proportion of the total variation in the response
variable has been explained by the model
- the closer to 1 the value of R2 is, the better "fit" of the regression line to the data
BRAINSCAPE1
, BRAINSCAPE1
R2 Interpretation - CORRECT ANSWER-If R2=.8872 you can said the variable explains
approximately 88.72% of the variation in the dependent variable.
Adjusted R2 - CORRECT ANSWER-- penalizes for # of variables used
- better indicator of model fit in multiple regression and can be used to inform decisions
Hypothesis test for a multiple variables - CORRECT ANSWER-Ho: B1=B2=B3=0
Ha: At least one coefficient is not equal to zero
Hypothesis test for a multiple variables - interpretation - CORRECT ANSWER-Ho: None
of the explanatory variables are useful
Ha: At least one of the explanatory variables are useful
Hypothesis test for a multiple variables - test - CORRECT ANSWER-F test
F-Test - CORRECT ANSWER-- global test assessing the overall fit of regression (not
specific variables)
- large F means small P
- Small F means large P
Hypothesis test for a reduced model - CORRECT ANSWER-Ho: Reduced model
variables = 0
Ha: At least one of the reduced model variables is not equal to zero
Hypothesis test for a reduced model - interpretation - CORRECT ANSWER-Ho: chose
the reduced model
Ha: at least one of the variables is contributing to the explanation of ht variation in y and
the full model is chosen as superior to the reduced
Hypothesis test for a reduced model - test - CORRECT ANSWER-partial F-test
Confidence Interval - CORRECT ANSWER-- an interval estimate of the average
predicted value for a number of observations
- "average" for "all"
- narrower
Prediction Interval - CORRECT ANSWER-- interval estimate of a predicted value for a
single observation
- "single" territory
- wider
Factors that influence CI and PI - CORRECT ANSWER-- sample size (larger =
narrower)
- spread of data (smaller error = narrower)
- confidence interval (99% is wider than 90%)
BRAINSCAPE1
