ECO 391 Final Exam questions and
answers graded A+
A regression's R2 can never be smaller than its adjusted R2 - correct answer ✔✔True
The sum of Squared Residuals (SSE) is the variation in the data that is explained by our model. -
correct answer ✔✔False
A 95% confidence interval is larger (has a bigger range) than a 90% confidence interval all else
constant. - correct answer ✔✔True
In the simple regression model, X is a response or dependent variable. - correct answer
✔✔False
If the null hypothesis is H0: ß1 > 0, then we are doing a right tailed hypothesis test. - correct
answer ✔✔False
The Standard Normal Distribution has a mean, median, and mode of 0. - correct answer
✔✔True
As our F-statistic decreases closer to zero, we are more likely to reject the null hypothesis. -
correct answer ✔✔False
Standard Deviation is a measure of the variation of the data series. - correct answer ✔✔True
A survey that only asks democrats "Who do you think will win the local election" is an example
of non-response bias. - correct answer ✔✔False
,Heteroskedasticity is when your X variable is correlated with your Y variable. - correct answer
✔✔False
Stratified Sampling is generally more accurate than cluster sampling and is preferred when
possible - correct answer ✔✔True
In a simple linear regression, if we can reject the null hypothesis of an F-test, we can also be
certain that our only independent variable is statistically significant. - correct answer ✔✔True
The sample variance can be positive or negative - correct answer ✔✔False
Imagine that the length of a newborn dragon, X, is normally distributed with a mean μ = 6
inches and standard deviation σ = 1.5 inches. What is the probability that the baby dragon will
be smaller than 3 inches? - correct answer ✔✔0.0228
Target sells a coat for $200 normally. You have a $10 off coupon and they are running a 20% off
sale. If the cashier takes the coupon off first before applying the sale, how much will the coat
cost (pre-tax)? - correct answer ✔✔$152
In order to test the joint significance of our model as a whole, which of the following would we
do? - correct answer ✔✔Run the regression and interpret the significance of the F-statistic.
Which of the following is true about a mode?
a. There can only be one mode for each data series.
b. The mode is always equal to the zero in the normal distribution.
c. The mode is the same as the 50th percentile for the dataset.
d. A dataset can have multiple modes only if there is more than one observation. - correct
answer ✔✔d. A dataset can have multiple modes only if there is more than one observation.
, If we run a simple linear regression and find an R2 = 0.45, we can conclude which of the
following?
a. There is a weak relationship between X and Y.
b. The variation in X explains 45% of the variation in Y.
c. There is a non-linear relationship between X and Y.
d. The variation in y explains 45% of the variation in X. - correct answer ✔✔b. The variation in X
explains 45% of the variation in Y
After you graph the data, you realize that there is a large left tail. Knowing this, which value will
be larger: the mean or median? - correct answer ✔✔Median
Which of the following will increase the width of a confidence interval?
a. Decrease the confidence level
b. Decrease the sample size
c. Decrease the margin of error
d. Increase the sample size - correct answer ✔✔b. Decrease the sample size
Which statement about homoscedasticity is correct?
a. A key assumption of the classic regression model (OLS) is that there is nohomoskedasticity.
b. Homoskedasticity is when the standard errors do not depend on the values of the X's.
c. Tests for significance are biased and inaccurate if there is homoskedasticity d. When
homoskedasticity is present, standard errors depend on any independent variable - correct
answer ✔✔b. Homoskedasticity is when the standard errors do not depend on the values of the
X's.
Suppose you run a multiple regression model with 4 independent variables. If there are 100
observations and the SSE = 75, what is the MSE? - correct answer ✔✔0.789
answers graded A+
A regression's R2 can never be smaller than its adjusted R2 - correct answer ✔✔True
The sum of Squared Residuals (SSE) is the variation in the data that is explained by our model. -
correct answer ✔✔False
A 95% confidence interval is larger (has a bigger range) than a 90% confidence interval all else
constant. - correct answer ✔✔True
In the simple regression model, X is a response or dependent variable. - correct answer
✔✔False
If the null hypothesis is H0: ß1 > 0, then we are doing a right tailed hypothesis test. - correct
answer ✔✔False
The Standard Normal Distribution has a mean, median, and mode of 0. - correct answer
✔✔True
As our F-statistic decreases closer to zero, we are more likely to reject the null hypothesis. -
correct answer ✔✔False
Standard Deviation is a measure of the variation of the data series. - correct answer ✔✔True
A survey that only asks democrats "Who do you think will win the local election" is an example
of non-response bias. - correct answer ✔✔False
,Heteroskedasticity is when your X variable is correlated with your Y variable. - correct answer
✔✔False
Stratified Sampling is generally more accurate than cluster sampling and is preferred when
possible - correct answer ✔✔True
In a simple linear regression, if we can reject the null hypothesis of an F-test, we can also be
certain that our only independent variable is statistically significant. - correct answer ✔✔True
The sample variance can be positive or negative - correct answer ✔✔False
Imagine that the length of a newborn dragon, X, is normally distributed with a mean μ = 6
inches and standard deviation σ = 1.5 inches. What is the probability that the baby dragon will
be smaller than 3 inches? - correct answer ✔✔0.0228
Target sells a coat for $200 normally. You have a $10 off coupon and they are running a 20% off
sale. If the cashier takes the coupon off first before applying the sale, how much will the coat
cost (pre-tax)? - correct answer ✔✔$152
In order to test the joint significance of our model as a whole, which of the following would we
do? - correct answer ✔✔Run the regression and interpret the significance of the F-statistic.
Which of the following is true about a mode?
a. There can only be one mode for each data series.
b. The mode is always equal to the zero in the normal distribution.
c. The mode is the same as the 50th percentile for the dataset.
d. A dataset can have multiple modes only if there is more than one observation. - correct
answer ✔✔d. A dataset can have multiple modes only if there is more than one observation.
, If we run a simple linear regression and find an R2 = 0.45, we can conclude which of the
following?
a. There is a weak relationship between X and Y.
b. The variation in X explains 45% of the variation in Y.
c. There is a non-linear relationship between X and Y.
d. The variation in y explains 45% of the variation in X. - correct answer ✔✔b. The variation in X
explains 45% of the variation in Y
After you graph the data, you realize that there is a large left tail. Knowing this, which value will
be larger: the mean or median? - correct answer ✔✔Median
Which of the following will increase the width of a confidence interval?
a. Decrease the confidence level
b. Decrease the sample size
c. Decrease the margin of error
d. Increase the sample size - correct answer ✔✔b. Decrease the sample size
Which statement about homoscedasticity is correct?
a. A key assumption of the classic regression model (OLS) is that there is nohomoskedasticity.
b. Homoskedasticity is when the standard errors do not depend on the values of the X's.
c. Tests for significance are biased and inaccurate if there is homoskedasticity d. When
homoskedasticity is present, standard errors depend on any independent variable - correct
answer ✔✔b. Homoskedasticity is when the standard errors do not depend on the values of the
X's.
Suppose you run a multiple regression model with 4 independent variables. If there are 100
observations and the SSE = 75, what is the MSE? - correct answer ✔✔0.789
