MENU
An analyst knows that the categorical predictor, zip_code, is an important predictor of a binary target. However, zip_code has too many levels to be a feasible predictor in a model. The analyst uses PROC CLUSTER to implement Greenacre's method to reduce the number of categorical levels.
What is the correct application of Greenacre's method in this situation?
Suggested Answer: D
Get Premium A00-240 Questions
as Interactive Web-Based Practice Test or PDF
Real Exam Format
Detailed Explanations
Instant Access
Get Your Practice Test
49 comments · newest first
BR
Brandon
11 months ago
D sounds complicated, but could work if done right!
AB
Abel
11 months ago
C is interesting, but I think A is more relevant here.
WI
Willow
12 months ago
Wait, can you really cluster using just zip_code values?
FR
Frederica
12 months ago
I disagree, B seems more straightforward.
SO
Solange
12 months ago
A is the way to go! Target proportions matter.
AR
Arlette
1 year ago
I vaguely recall that using dummy variables could complicate the clustering process, so I'm leaning towards option A for this one.
LI
Linn
1 year ago
I practiced a similar question where we clustered based on case counts, but I think that might not be the best approach here either.
LE
Leoma
1 year ago
I'm not entirely sure, but I feel like clustering based on the zip_code values themselves might not capture the target's influence effectively.
JO
Johnson
1 year ago
I remember discussing how Greenacre's method focuses on the relationships between categories, so I think using the target proportion for each zip_code makes sense.
RE
Reuben
1 year ago
I'm pretty confident that the correct answer is A. Clustering the levels using the target proportion for each zip_code makes the most sense to reduce the number of levels while preserving the predictive power of the variable.
ES
Essie
1 year ago
I'm a bit confused on this one. Clustering the levels using the zip_code values or the number of cases doesn't seem quite right. Dummy coding the zip_code levels also doesn't seem relevant here.
JA
Jamal
1 year ago
Okay, the key here is that we're trying to reduce the number of levels in the zip_code predictor. I think using the target proportion for each zip_code as the input to the clustering makes the most sense.
DE
Dell
1 year ago
This looks like a tricky question. I'll need to think carefully about the different options and how Greenacre's method works.
OL
Olene
1 year ago
Hmm, this seems like a tricky one. I'll need to carefully review the options and think through the implications of each setting.
DI
Dick
1 year ago
This seems straightforward - I think the answer is to assign the existing service appointment to the crew.
JO
Johnetta
1 year ago
I remember calculating probabilities before, but I'm not completely sure how to approach this one with different ball colors.
MA
Matthew
1 year ago
I don't know, guys. Greenacre's method sounds a bit like rocket science to me. Maybe we should just ask the professor for the answer key.
CA
Carrol
1 year ago
D) Clustering the levels using dummy coded zip_code levels as inputs.
CU
Curt
1 year ago
C) Clustering the levels using the number of cases in each zip_code as input.
MA
Marion
1 year ago
B) Clustering the levels using the zip_code values as input.
RA
Rana
1 year ago
A) Clustering the levels using the target proportion for each zip_code as input.
IS
Isaac
1 year ago
Ah, the number of cases in each zip_code, the classic approach. I like it, nice and straightforward.
AS
Ashlee
1 year ago
C) Clustering the levels using the number of cases in each zip_code as input.
OL
Olive
1 year ago
B) Clustering the levels using the zip_code values as input.
TI
Tish
1 year ago
A) Clustering the levels using the target proportion for each zip_code as input.
SA
Sage
1 year ago
I'm not sure about using dummy coded zip_code levels as inputs. Doesn't that seem a bit like overkill for this problem?
MA
Margo
1 year ago
C) Clustering the levels using the number of cases in each zip_code as input.
SH
Shenika
1 year ago
I agree, using dummy coded zip_code levels might be too complex for this situation.
LE
Leah
1 year ago
I think using the target proportion for each zip_code makes more sense.
VI
Vincent
1 year ago
B) Clustering the levels using the zip_code values as input.
AL
Alishia
1 year ago
B) Clustering the levels using the zip_code values as input.
YO
Youlanda
1 year ago
A) Clustering the levels using the target proportion for each zip_code as input.
LI
Ligia
1 year ago
A) Clustering the levels using the target proportion for each zip_code as input.
WI
Willetta
2 years ago
Clustering the levels using the zip_code values as input? That's a bold move, Cotton. Let's see if it pays off.
SE
Sena
1 year ago
D) Clustering the levels using dummy coded zip_code levels as inputs.
CA
Caitlin
1 year ago
C) Clustering the levels using the number of cases in each zip_code as input.
MA
Maricela
1 year ago
B) Clustering the levels using the zip_code values as input.
LE
Lenna
1 year ago
A) Clustering the levels using the target proportion for each zip_code as input.
WI
William
2 years ago
I'm not sure, but I think C) Clustering the levels using the number of cases in each zip_code as input could also be a valid approach.
TA
Tabetha
2 years ago
Hmm, I think option A is the way to go. Using the target proportion for each zip_code as input seems like the most logical approach to cluster the levels.
ON
Oneida
1 year ago
I think so too. It's important to consider the target proportion when clustering the levels.
AL
Alesia
1 year ago
I think so too. It's important to consider the target proportion when clustering the levels.
BE
Bernadine
1 year ago
I agree, option A makes the most sense. It takes into account the target proportion for each zip_code.
EN
Ena
1 year ago
I agree, option A makes the most sense. It takes into account the target proportion for each zip_code.
MA
Malinda
2 years ago
I disagree, I believe it should be D) Clustering the levels using dummy coded zip_code levels as inputs.
RO
Rolande
2 years ago
I think the correct application is A) Clustering the levels using the target proportion for each zip_code as input.
TY
Ty
2 years ago
I agree with Layla, A) makes more sense because we want to reduce the number of levels based on the target variable.
EM
Emily
2 years ago
I disagree, I believe it should be D) Clustering the levels using dummy coded zip_code levels as inputs.
LA
Layla
2 years ago
I think the correct application is A) Clustering the levels using the target proportion for each zip_code as input.