01
Choose one
1 point
A model predicts home price as $50,000 + $250 × area in square feet. What price does it predict for a home with 1,000 square feet?
Free Practice Quiz Question List
4 Regression Online Quiz Questions
Use this free practice quiz with 20 questions to review 4 Regression, test your knowledge, and prepare for your next test or exam.
20 questions
02
True or false
1 point
A regression model uses labeled examples to predict a numeric target.
- A
True
- B
False
03
Written response
1 point
In a labeled example, what is the term for an input value used by a regression model to predict the value of interest?
04
Fill in the blank
1 point
Ordinary least squares chooses coefficients to minimize the between actual and predicted targets.
05
Choose one
1 point
A home-price model has an intercept of $50,000 and uses floor area as its only feature. Which interpretation of the intercept is most accurate?
- A
It is the predicted change in price for each additional square foot.
- B
It is the average price of every home in the dataset.
- C
It is the predicted price at zero square feet, which may have little practical meaning if zero is outside the useful data range.
- D
It proves that increasing floor area causes the home's price to rise.
06
Written response
1 point
What is the statistical term for the actual target minus the predicted target?
07
Choose all
1 point
Select all accurate statements about interpreting a coefficient in a multiple-feature regression model.
- A
It describes the model's predicted change for that feature while the other included features are held fixed.
- B
It measures the feature's effect while all other features are ignored.
- C
It describes a model relationship but does not by itself prove that changing the feature causes the target to change.
- D
It guarantees that the feature will have the same relationship with the target in every population.
08
Fill in the blank
1 point
Fit the regression model using only the set, keeping the test set separate for evaluation.
09
Choose one
1 point
A regression model has an R² value below zero on a test set. What does this indicate?
- A
The model predicts every target perfectly.
- B
The model performs worse than predicting the test-set mean.
- C
The model's typical error is exactly zero.
- D
The model performs exactly as well as predicting the test-set mean.
10
Choose all
1 point
Select all accurate statements about regression evaluation metrics.
- A
MAE is the average signed error, so positive and negative errors cancel.
- B
MAE is expressed in the target's units.
- C
RMSE penalizes large errors more strongly than MAE.
- D
R² alone does not show the typical size of an error.
11
Choose one
1 point
A project especially wants to penalize occasional very large prediction errors. Which metric has the stated property that makes it especially relevant to consider?
- A
Mean absolute error, because it gives large errors a stronger penalty than root mean squared error.
- B
R², because it reports the typical error in the target's units.
- C
Root mean squared error, because it penalizes large errors more strongly than mean absolute error.
- D
The intercept, because it measures the average prediction error.
12
Choose one
1 point
A home-price model has a positive residual for a particular home. What does this tell you about the model's prediction for that home?
- A
The model predicted too low.
- B
The model predicted too high.
- C
The model's prediction was exactly correct.
- D
The prediction error must have been caused by changing the feature.
13
Choose one
1 point
A linear model's intercept represents a prediction at a feature value far outside the range observed in the training data. Which conclusion is most justified?
- A
The model must be discarded whenever its intercept is not zero.
- B
The intercept is the average of the observed targets.
- C
The intercept is the prediction at zero feature value, but that prediction may have little practical meaning if zero is outside the useful data range.
- D
The intercept proves what would happen if the feature were changed to zero.
14
Choose one
1 point
A transport team wants a model to predict the number of minutes each trip will take. Which approach fits this goal?
- A
Classifying a trip as local or long-distance
- B
Predicting how many minutes a trip will take
- C
Sorting trips by driver surname
- D
Grouping trips by route name
15
Choose one
1 point
A team has labeled examples and wants to estimate how well its regression model performs on examples it did not train on. Which procedure best supports that goal?
- A
Fit on the test set and evaluate on the training set
- B
Fit on both sets, then evaluate on the test set
- C
Fit on the training set and reserve the test set for evaluation
- D
Fit and evaluate on the test set, using the training set only to choose features
16
True or false
1 point
If a regression model fits its training examples extremely well, it is guaranteed to perform well on unseen examples.
- A
True
- B
False
17
Written response
1 point
In the example code, which model method learns the parameters from the training data?
18
Written response
1 point
A home-price model predicts a price by adding $50,000 to $250 for each square foot of floor area. What price does it predict for a 1,000-square-foot home? Enter the amount in dollars.
19
True or false
1 point
A regression model has an R² of 0 on a test set. Its performance is equivalent to predicting the test-set mean. Is this statement true or false?
- A
True
- B
False
20
Open ended
1 point
A straight-line model predicts home prices from floor area. When you examine its residuals, they follow a consistent curve as floor area increases. What might this pattern indicate about the model, and what would be a sensible next step?