Free Practice Quiz Question List

6 Data Fitting and Empirical Models Online Quiz Questions

Use this free practice quiz with 20 questions to review 6 Data Fitting and Empirical Models, test your knowledge, and prepare for your next test or exam.

20 questions

Choose one

1 point

What is the main reason to inspect a scatterplot before fitting an empirical model?

A
To prove that one variable causes the other
B
To inspect the observed relationship before choosing a model
C
To guarantee that the fitted model will predict accurately outside the data range
D
To calculate the exact measurement error for every observation

Written response

1 point

What term describes the observed response minus the fitted response for one observation?

Fill in the blank

1 point

The most common coefficient-estimation method in simple linear regression is , which minimizes the .

Choose one

1 point

For the fitted equation $\hat y=18.2+2.7x$ , what does $2.7$ represent?

A
The predicted response when $x=0$
B
The average residual of the fitted model
C
The estimated change in the response for a one-unit increase in $x$
D
The fraction of response variation accounted for by the model

Choose all

1 point

Which residual-plot patterns are warning signs that the fitted model may need investigation? Select all that apply.

A
A curved pattern
B
An approximately random cloud centered around zero
C
A funnel shape
D
An isolated large residual

Written response

1 point

A cyclist's fitted distance model is $\hat d=12.4+4.8t$ , where $t$ is measured in hours. What distance does the model predict after $3.5$ hours? Enter the value in miles.

Fill in the blank

1 point

The is calculated by adding the squared residuals, $\sum e_i^2$ .

Choose one

1 point

Which situation best describes overfitting?

A
A model has too few observations to calculate a residual
B
A model uses a scatterplot before fitting
C
A model has a high residual standard deviation in the response units
D
A model follows random fluctuations in the available data and performs poorly on new data

Choose all

1 point

Which considerations should be used when comparing candidate empirical models? Select all that apply.

A
Whether residual patterns show systematic problems
B
Whether typical prediction errors are acceptable
C
Whether the model has the smallest training error regardless of complexity
D
Whether the model predicts new observations well

Open ended

1 point

Explain how you would responsibly interpret and report predictions from an empirical model. Include the scope of the data, important uncertainty or limitations, and why a fitted association should not automatically be treated as a causal relationship.

Choose one

1 point

Why can taking logarithms be useful when fitting an exponential model $\hat y=ae^{kx}$ ?

A
It changes every exponential relationship into a polynomial relationship
B
It can produce a relationship linear in $x$ , allowing linear regression on transformed data
C
It guarantees that predictions on the original response scale are unbiased
D
It removes the need to examine residual behavior

Choose one

1 point

Which situation best illustrates overfitting in an empirical model?

A
The model has too few parameters to describe the data.
B
The model follows random fluctuations in the available data and performs poorly on new data.
C
The model uses a scatterplot before fitting.
D
The model has residuals measured in the same units as the response.

Choose one

1 point

A residual plot shows a funnel shape as the explanatory variable increases. What issue does this pattern most directly suggest?

A
The relationship is perfectly linear.
B
The observations are necessarily independent.
C
The variability changes as the predictor changes.
D
The response has no unexplained variation.

Choose one

1 point

For the fitted model $hat y=18.2+2.7x$ , what does the slope imply about a one-unit increase in $x$ ?

A
The predicted response increases by 2.7 units.
B
The predicted response increases by 18.2 units.
C
The predicted response equals 2.7 when the explanatory variable is zero.
D
The predicted response decreases by 18.2 units.

Choose one

1 point

Why is a high $R^2$ not sufficient by itself to validate a fitted model?

A
It proves that the explanatory variable causes the response to change.
B
It guarantees accurate predictions outside the observed data range.
C
It shows that all residual assumptions are satisfied.
D
It may coexist with systematic residual patterns or poor performance outside the observed range.

True or false

1 point

True or false: A scatterplot that shows an association between two variables by itself proves that changing one variable causes the other to change.

A
True
B
False

Written response

1 point

What term describes a regression situation in which the variability of the residuals changes with the level of the predictor?

True or false

1 point

A model fitted using observations with x-values from 10 to 20 may produce highly unreliable predictions at x=50, even if it fits the observed data well. True or false?

A
True
B
False

True or false

1 point

A nonrepresentative sample can limit how well an empirical model generalizes, even when the model fits the observed data well. True or false?

A
True
B
False

Written response

1 point

A model was fitted using data for x-values from 2 to 8. What is the statistical term for using that model to predict y when x=15?