ISYE 6501 -Exam 2 Wks 8 - 12
Study online at https://quizlet.com/_dhnu3f
1. Building simpler models with A. Overfitting
fewer factors helps avoid D. Difficulty of interpretation
which problems?
A. Overfitting
B. Low prediction quality
C. Bias in the most important
factors
D. Difficulty in interpretation
2. Two main reasons to limit # of 1. Overfitting
factors in a model. 2. Simplicity
3. When is overfitting likely to When the number of factors is close to the number of data
happen? points.
4. How does using a # of factors The model too closely fits the random efffects. It fits that data
that is close to the number of set well, but fails to predict well on a new data set.
data points cause overfitting?
5. Three reasons simple models 1. Less data is required
are better than complex mod- 2. Less chance of including insignificant factors
els. 3. Easier to interpret
6. Which of these is a key differ- A. If the data isn't scaled, the coefficients can have artificially
ence between stepwise regres- different orders of magnitude, which means they'll have un-
sion and lasso regression? balanced effects on the lasso constraint.
A. Lasso regression requires
the data to first be scaled
B. Stepwise regression gives
many models to choose from,
while lasso gives just one.
7.
, ISYE 6501 -Exam 2 Wks 8 - 12
Study online at https://quizlet.com/_dhnu3f
Name three greedy approach- 1. Forward Selection
es to variable selection. 2. Backward Elimination
3. Stepwise Regression
8. Name two global approaches 1. Lasso
to variable selection. 2. Elastic Net
9. What does a greedy approach At each step, the model does one thing that looks best with-
to variable selection mean? out taking future options into consideration. A more classical
approach.
10. How does forward selection Forward selection starts with zero factors and backward elim-
differ from backward elimina- ination starts with all factors.
tion?
11. What is stepwise regression? A combination of forward selection and backward elimination.
At each step, a variable is considered for addition or elimina-
tion based on some prespecified criterion.
12. Name two approaches to use 1. Start with all factors
with stepwise regression. 2. Start with no factors
13. What restriction does the lasso The sum of the coefficients can't be too large. "t"
approach add?
14. How does lasso use the t value? It uses that on the most important coefficients and the others
will be zero so those factors won't be part of the model.
15. What do you need to with Scale the data.
the data whenever you're con-
straining the sum of coeffi-
cients?
16.
Study online at https://quizlet.com/_dhnu3f
1. Building simpler models with A. Overfitting
fewer factors helps avoid D. Difficulty of interpretation
which problems?
A. Overfitting
B. Low prediction quality
C. Bias in the most important
factors
D. Difficulty in interpretation
2. Two main reasons to limit # of 1. Overfitting
factors in a model. 2. Simplicity
3. When is overfitting likely to When the number of factors is close to the number of data
happen? points.
4. How does using a # of factors The model too closely fits the random efffects. It fits that data
that is close to the number of set well, but fails to predict well on a new data set.
data points cause overfitting?
5. Three reasons simple models 1. Less data is required
are better than complex mod- 2. Less chance of including insignificant factors
els. 3. Easier to interpret
6. Which of these is a key differ- A. If the data isn't scaled, the coefficients can have artificially
ence between stepwise regres- different orders of magnitude, which means they'll have un-
sion and lasso regression? balanced effects on the lasso constraint.
A. Lasso regression requires
the data to first be scaled
B. Stepwise regression gives
many models to choose from,
while lasso gives just one.
7.
, ISYE 6501 -Exam 2 Wks 8 - 12
Study online at https://quizlet.com/_dhnu3f
Name three greedy approach- 1. Forward Selection
es to variable selection. 2. Backward Elimination
3. Stepwise Regression
8. Name two global approaches 1. Lasso
to variable selection. 2. Elastic Net
9. What does a greedy approach At each step, the model does one thing that looks best with-
to variable selection mean? out taking future options into consideration. A more classical
approach.
10. How does forward selection Forward selection starts with zero factors and backward elim-
differ from backward elimina- ination starts with all factors.
tion?
11. What is stepwise regression? A combination of forward selection and backward elimination.
At each step, a variable is considered for addition or elimina-
tion based on some prespecified criterion.
12. Name two approaches to use 1. Start with all factors
with stepwise regression. 2. Start with no factors
13. What restriction does the lasso The sum of the coefficients can't be too large. "t"
approach add?
14. How does lasso use the t value? It uses that on the most important coefficients and the others
will be zero so those factors won't be part of the model.
15. What do you need to with Scale the data.
the data whenever you're con-
straining the sum of coeffi-
cients?
16.
