Thank you Mark & Dunlap,
Will make changes to the variable as suggested. Thank you for your time &
assistance.
On Wed, Aug 31, 2016 at 9:16 PM, Mark Sharp wrote:
> Shivi,
>
> Looking at the help from ?WOE, ?WOETable, and ?IV, your Y vector in all
> cases is to be categorical and it is numeric.
Shivi,
Looking at the help from ?WOE, ?WOETable, and ?IV, your Y vector in all cases
is to be categorical and it is numeric.
Mark
R. Mark Sharp, Ph.D.
Director of Primate Records Database
Southwest National Primate Research Center
Texas Biomedical Research Institute
P.O. Box 760549
San Antonio
I see the printout (not an "error") that you describe on the 2nd example
you gave:
> t2 <- WOETable(X=SFDC1$case_age, Y=SFDC1$survey)
> print(t2)
[1] GOODS BADS TOTAL PCT_G PCT_B WOE IV
<0 rows> (or 0-length row.names)
The result of WOETable is perfectly legal - it is a data.frame with no rows
These are the packages i am using:
library(woe) #WEIGHT OF EVIDENCE
library(InformationValue) #INFORMATION VALUE
The syntax used is :
WOE(X=SFDC1$log_caseage, Y=SFDC1$survey)
WOETable(X=SFDC1$case_age, Y=SFDC1$survey)
IV(X=SFDC1$case_age, Y=SFDC1$survey)
On Wed, Aug 31, 2016 at 12:54 AM, Mark S
You need to show what R expressions you ran before running into this
problem, including calls to library() or require().
Bill Dunlap
TIBCO Software
wdunlap tibco.com
On Tue, Aug 30, 2016 at 12:15 PM, Shivi Bhatia wrote:
> Hi Mark,
> What i understand, probably when i run the WOE & IV to check s
Shivi,
What package(s) are you using. What functions are you using. How are you
calling the functions. A reproducible sample has all of the actual code needed
to create a representative error. There are multiple packages you could be
using to look at weight of evidence and information value. Fo
Hi Mark,
What i understand, probably when i run the WOE & IV to check significant
variables that is where i get this error. Thanks for your assistance Mark
really appreciate i will look into some other measure on this.
On Wed, Aug 31, 2016 at 12:35 AM, Mark Sharp wrote:
> Shivi,
>
> It is likely
Shivi,
It is likely that William knows what you are trying to do because of his
considerable experience, but I am not able to figure it out from what you have
written. You have apparently sent the output from something like
dput(SFDC[1:50, ]), but I still do not know what you did to get the err
Hi William/ Mark,
I am using WOE & IV (weight of evidence) reduce the number of independent
vars.
I have read this data as a csv file.
reproducible example for your reference please:
structure(list(date = structure(c(6L, 6L, 6L, 6L, 6L, 6L, 14L,
14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 14L, 1
You did not say what operation gave you the error.
I can get that message (which is not an "error") if I print
an illegally constructed data.frame, one without the
row.names attribute.
> illegalDF <- structure(class="data.frame", list(ColumnA = 1:3))
> illegalDF
[1] ColumnA
<0 rows> (or 0-length
Shivi,
Can you show the code that throws the error?
<0 rows> (or 0-length row.names)
Of course as always a reproducible sample would be great. Perhaps you can make
a small subset of the data and use dput() to provide a defined object.
R. Mark Sharp, Ph.D.
Director of Primate Records Database
So
Hi Mark,
It gives me num [1:5083]. I have used head also to see first 10 obs:
head(SFDC$case_age,10)
[1] 24.84 0.05 13.38 0.15 11.11 4.16 8.13 0.07 3.61 0.00
Thanks.
On Tue, Aug 30, 2016 at 10:05 PM, Mark Sharp wrote:
> What do you get from
> str(SFDC$case_age)
>
> Mark
>
> R. Mark Sh
What do you get from
str(SFDC$case_age)
Mark
R. Mark Sharp, Ph.D.
Director of Primate Records Database
Southwest National Primate Research Center
Texas Biomedical Research Institute
P.O. Box 760549
San Antonio, TX 78245-0549
Telephone: (210)258-9476
e-mail: msh...@txbiomed.org
> On Aug 30,
I know this question has been asked zillion times but even after consulting
Stack Overflow & other forum cant figure out the reason.
I have one var in my data-set names case age. This variable is numeric as:
class(SFDC$case_age)
*numeric*
however it throws this error:
<0 rows> (or 0-length row
14 matches
Mail list logo
