SCMA 350 EXAM QUESTIONS AND ANSWERS
In regression analysis, the variables used to help explain or predict the response
variable are referred to as which type of variables? - Answer -Independent
In regression analysis, the variable we are trying to explain or predict is described as
which type of variable? - Answer -Dependent
In regression analysis, if there are several explanatory variables, it is referred to as -
Answer -multiple regression.
In regression analysis, which of the following causal relationships are possible? -
Answer -All of these options are possible.
Correlation is a summary measure that indicates - Answer -the strength of the linear
relationship between pairs of variables.
A correlation value of zero indicates - Answer -no linear relationship.
How do you determine whether a regression model provides useful predictions? -
Answer -Analyze the residuals versus the fitted values
How can you validate that a regression model predicts well on new data? - Answer -By
using a training and validation split of the original dataset
Residual plots during regression analysis help assess which assumption? - Answer -
Normality of residuals
An error term represents the vertical distance from any point to the - Answer -
population regression line.
In regression analysis, multicollinearity refers to the - Answer -explanatory variables
being highly correlated.
Determining which variables to include in regression analysis by estimating a series of
regression equations by successively adding or deleting variables according to
prescribed rules is referred to as which type of regression? - Answer -Stepwise
The objective typically used in the tree types of equation-building procedures is to find
the equation with - Answer -a small se and a large R2.
Forward regression - Answer -begins with no explanatory variables in the equation and
successively adds one at a time until no remaining variables make a significant
contribution.
, How does a t-value in regression output provide the basis for a hypothesis test? -
Answer -The t-value shows how many standard errors the coefficient is from zero.
What is indicated when the p-value for an explanatory variable in a regression model is
small (less than 0.05)? - Answer -The coefficient for that variable is statistically
significant.
What does a large t-value for a regression coefficient suggest regarding that coefficient?
- Answer -It has a statistically significant nonzero value.
How does the t-statistic allow you to test whether a regression coefficient equals zero? -
Answer -It measures the coefficient relative to its standard error.
A large t-value for a regression coefficient suggests that - Answer -it is statistically
significant.
In linear regression, we fit the least squares line to a set of values (or points on a
scatterplot). The distance from the line to a point is called the - Answer -residual
In choosing the "best-fitting" line through a set of points in linear regression, we choose
the one with the - Answer -smallest sum of squared residuals.
The standard error of the estimate Se is essentially the - Answer -standard deviation of
the residuals.
The adjusted R2 adjusts R2 for - Answer -the number of explanatory variables in a
multiple regression model.
The R2 value in regression analysis measures what? - Answer -The percentage of
variation in the dependent variable explained by the regression model
Which regression output value indicates the percentage of variation in the dependent
variable that is unexplained by the model? - Answer -1 - R2
Based on a regression analysis, the R 2 is 0.72. What does this value indicate? -
Answer -About 72% of the variation in the dependent variable is explained.
Based on a regression analysis, the R 2 is 0.84. What does this value indicate? -
Answer -About 84% of the variation in the dependent variable is explained.
Which of the following best describes multicollinearity in multiple regression analysis? -
Answer -Correlated explanatory variables that make their individual effects difficult to
estimate
How can you detect multicollinearity in a multiple regression model? - Answer -By
inspecting the correlation matrix
In regression analysis, the variables used to help explain or predict the response
variable are referred to as which type of variables? - Answer -Independent
In regression analysis, the variable we are trying to explain or predict is described as
which type of variable? - Answer -Dependent
In regression analysis, if there are several explanatory variables, it is referred to as -
Answer -multiple regression.
In regression analysis, which of the following causal relationships are possible? -
Answer -All of these options are possible.
Correlation is a summary measure that indicates - Answer -the strength of the linear
relationship between pairs of variables.
A correlation value of zero indicates - Answer -no linear relationship.
How do you determine whether a regression model provides useful predictions? -
Answer -Analyze the residuals versus the fitted values
How can you validate that a regression model predicts well on new data? - Answer -By
using a training and validation split of the original dataset
Residual plots during regression analysis help assess which assumption? - Answer -
Normality of residuals
An error term represents the vertical distance from any point to the - Answer -
population regression line.
In regression analysis, multicollinearity refers to the - Answer -explanatory variables
being highly correlated.
Determining which variables to include in regression analysis by estimating a series of
regression equations by successively adding or deleting variables according to
prescribed rules is referred to as which type of regression? - Answer -Stepwise
The objective typically used in the tree types of equation-building procedures is to find
the equation with - Answer -a small se and a large R2.
Forward regression - Answer -begins with no explanatory variables in the equation and
successively adds one at a time until no remaining variables make a significant
contribution.
, How does a t-value in regression output provide the basis for a hypothesis test? -
Answer -The t-value shows how many standard errors the coefficient is from zero.
What is indicated when the p-value for an explanatory variable in a regression model is
small (less than 0.05)? - Answer -The coefficient for that variable is statistically
significant.
What does a large t-value for a regression coefficient suggest regarding that coefficient?
- Answer -It has a statistically significant nonzero value.
How does the t-statistic allow you to test whether a regression coefficient equals zero? -
Answer -It measures the coefficient relative to its standard error.
A large t-value for a regression coefficient suggests that - Answer -it is statistically
significant.
In linear regression, we fit the least squares line to a set of values (or points on a
scatterplot). The distance from the line to a point is called the - Answer -residual
In choosing the "best-fitting" line through a set of points in linear regression, we choose
the one with the - Answer -smallest sum of squared residuals.
The standard error of the estimate Se is essentially the - Answer -standard deviation of
the residuals.
The adjusted R2 adjusts R2 for - Answer -the number of explanatory variables in a
multiple regression model.
The R2 value in regression analysis measures what? - Answer -The percentage of
variation in the dependent variable explained by the regression model
Which regression output value indicates the percentage of variation in the dependent
variable that is unexplained by the model? - Answer -1 - R2
Based on a regression analysis, the R 2 is 0.72. What does this value indicate? -
Answer -About 72% of the variation in the dependent variable is explained.
Based on a regression analysis, the R 2 is 0.84. What does this value indicate? -
Answer -About 84% of the variation in the dependent variable is explained.
Which of the following best describes multicollinearity in multiple regression analysis? -
Answer -Correlated explanatory variables that make their individual effects difficult to
estimate
How can you detect multicollinearity in a multiple regression model? - Answer -By
inspecting the correlation matrix
