Free Practice Quiz Question List

7 Regression for Business Analysis Online Quiz Questions

Use this free practice quiz with 20 questions to review 7 Regression for Business Analysis, test your knowledge, and prepare for your next test or exam.

20 questions

True or false

1 point

A regression coefficient is statistically significant. Does that fact alone establish that changing its predictor causes the outcome to change?

A
True
B
False

True or false

1 point

Is approximate residual normality required in order to calculate least-squares coefficient estimates?

A
True
B
False

Fill in the blank

1 point

Ordinary least squares chooses model coefficients to minimize the .

Fill in the blank

1 point

A regression model with a squared predictor can still be linear in its .

Written response

1 point

What is the statistical term for the difference between an observed outcome and its predicted value?

Choose one

1 point

A retailer models weekly sales in thousands of dollars using advertising spend in thousands of dollars and price in dollars. In the fitted model, the advertising coefficient is 2.4. Holding price constant, how should the coefficient be interpreted for an additional $1,000 in advertising?

A
An estimated $2,400 increase in weekly sales
B
An estimated $1,100 decrease in weekly sales
C
An estimated $2.40 increase in weekly sales
D
An estimated $2,400 increase in weekly sales for every $1 increase in price

Written response

1 point

A retailer's fitted model is $\widehat{Sales}=12+2.4(Advertising)-1.1(Price)$ , where sales are in thousands of dollars, advertising is in thousands of dollars, and price is in dollars. What weekly sales value does the model predict when $Advertising=5$ and $Price=10$ ?

Choose one

1 point

A business wants to evaluate a regression model that forecasts future sales. Which train/test split is most appropriate for preventing future information from leaking into model fitting?

A
Use earlier observations for training and later observations for testing
B
Randomly mix future and past observations between training and testing
C
Evaluate only on the data used to fit the model
D
Choose the split that produces the highest test score, regardless of time order

Choose all

1 point

A business analyst inspects residual plots for a linear regression model. Select all conclusions that are supported by the diagnostic patterns described.

A
A curved residual pattern may indicate that the model is missing structure.
B
A fan-shaped residual spread may indicate nonconstant error variance.
C
Either pattern proves that the predictor causes the outcome.
D
A high overall fit statistic makes these residual patterns irrelevant.

Choose one

1 point

When a categorical predictor is represented by category indicators in a regression, what does a category coefficient typically express?

A
It is the outcome difference from the overall mean for every category.
B
It compares that category with a chosen reference category.
C
It is automatically the causal effect of belonging to that category.
D
It compares that category only with the category having the largest sample size.

Choose all

1 point

Select all accurate statements about regression performance measures.

A
R-squared is the percentage of predictions that are correct.
B
RMSE penalizes large errors more heavily than MAE does.
C
Adjusted R-squared accounts for the number of predictors and is mainly used to compare models fit to the same response and dataset.
D
A higher R-squared by itself establishes that a model is useful for a business decision.

Open ended

1 point

A business plans to use a regression model to predict sales at advertising levels far beyond those in the data used to fit the model. What is the main concern, and what should the business do before relying on that prediction?

Choose one

1 point

Which statement correctly distinguishes a confidence interval for a regression coefficient from a prediction interval for an individual future outcome?

A
A confidence interval gives plausible values for a coefficient; a prediction interval describes uncertainty for an individual future outcome.
B
A confidence interval gives the exact value of a future outcome; a prediction interval gives plausible coefficient values.
C
Both intervals describe plausible values for the same regression coefficient.
D
A prediction interval measures whether a coefficient is statistically significant; a confidence interval measures prediction accuracy.

True or false

1 point

True or false: A regression table by itself proves that a predictor caused the observed change in a business outcome.

A
True
B
False

Written response

1 point

A model predicts an outcome of 79 units, while the observed outcome is 84 units. What is the residual, calculated as observed minus predicted?

Written response

1 point

Which regression performance metric is the average absolute prediction error and is expressed in the outcome's units?

Choose one

1 point

A retailer estimates $\widehat{Sales}=12+2.4(Advertising)-1.1(Price)$ , where weekly sales and advertising spend are both measured in thousands of dollars and price is measured in dollars. Which interpretation of the advertising coefficient is appropriate?

A
An additional $1$ dollar of advertising is associated with a $2.4$ -dollar increase in weekly sales.
B
An additional $1$ thousand dollars in advertising is associated with a $2.4$ thousand dollar increase in weekly sales, holding price constant.
C
An additional $1$ thousand dollars in advertising guarantees a $2.4$ thousand dollar increase in weekly sales.
D
An additional $1$ dollar in price is associated with a $2.4$ thousand dollar increase in weekly sales.

Choose one

1 point

A business sees a high R-squared for a regression model evaluated on held-out data. What does this measure indicate?

A
The percentage of individual predictions that exactly match the observed outcomes.
B
The percentage reduction in the model's average absolute error.
C
The fraction of outcome variation explained by the fitted model in the evaluation data.
D
The probability that the model's predictors cause the outcome.

Choose one

1 point

According to the usual interpretation of regression intervals, which is usually wider: a confidence interval for a coefficient or a prediction interval for an individual future outcome?

A
The confidence interval for a coefficient
B
The prediction interval for an individual future outcome
C
Both intervals are usually the same width
D
Neither interval can be compared in width

Choose one

1 point

A regression includes a categorical predictor for customer plan type. How should a coefficient for one plan category typically be interpreted?

A
Each category coefficient typically represents a comparison with the chosen reference category.
B
Each category coefficient is the outcome value for that category, regardless of other predictors.
C
The reference category must be the category with the largest sample size.
D
Categorical predictors cannot be included in a regression model.