[R] help - hoslem.test

PIKAL Petr petr.pikal at precheza.cz
Mon May 4 15:24:40 CEST 2015


Hi

thanks for data.

na.exclude does not excludes NA values. Actually it computes result without considering NAs but keeps those NA values in propper positions so that the result has the same length as input

Instead of several ifelse you can use

q131 <- as.numeric(cut(id3$q13,c(0,1.5,5)))-1

fitted values start with NA values

> fitted(tp1)
        1241         1242         1243         1244         1245         1246
          NA           NA 2.143345e-11 2.143345e-11 2.143345e-11 2.143345e-11
        1247         1248         1256         1268
2.143345e-11 2.143345e-11 2.143345e-11 2.143345e-11

which obviously hoslem.test is not adapted to.

>
> #Error in quantile.default(yhat, probs = seq(0, 1, 1/g)) :
>  # missing values and NaN's not allowed if 'na.rm' is FALSE

you can use glm without na.action and your fitted values will be without NA.

> tp1 <- glm(q131 ~ q11 + q10+q12+edcat + q08+q06+ q14, family = binomial(link = "logit"), data=id3)
> fitted(tp1)
        1243         1244         1245         1246         1247         1248
2.143345e-11 2.143345e-11 2.143345e-11 2.143345e-11 2.143345e-11 2.143345e-11
        1256         1268
2.143345e-11 2.143345e-11

however
tp1$q131 does not exist

> tp1$q131
NULL
So again, test throws error.

Maybe you wanted

> hoslem.test(na.omit(id3$q131), fitted(tp1), g=10)

        Hosmer and Lemeshow goodness of fit (GOF) test

data:  na.omit(id3$q131), fitted(tp1)
X-squared = NaN, df = 8, p-value = NA

Warning message:
In Ops.factor(1, y) : ‘-’ not meaningful for factors

However q131 is factor (actually you changed it to factor)

If I change it to numeric

> id3$q131<-q131
> tp1 <- glm(q131 ~ q11 + q10+q12+edcat + q08+q06+ q14, family = binomial(link = "logit"), data=id3)
> hoslem.test(na.omit(id3$q131), fitted(tp1), g=10)

        Hosmer and Lemeshow goodness of fit (GOF) test

data:  na.omit(id3$q131), fitted(tp1)
X-squared = NaN, df = 8, p-value = NA

hoslem test works - at least it does not throw errors.

So best way for you would be to understand basic differences in object types (numeric, factor, ....) by reading R intro and think about what hoslem test needs as an input.

Cheers
Petr

> -----Original Message-----
> From: Luciane Maria Pilotto [mailto:lutipilotto at yahoo.com.br]
> Sent: Monday, May 04, 2015 2:46 PM
> To: r-help at r-project.org; PIKAL Petr
> Subject: RE: [R] help - hoslem.test
>
> Hi
>
> I'm trying to make a reproducible example using the command "dput" as
> fallows.
>
> The problem occurs when running the test of Hosmer and Lemeshow
> (hoslem.test) for residuals gives error. I'm using the command
> "na.action" to exclude the NA values.
>
> Thanks,
> Luciane
>
> ###############################################
>
> load(file.choose())#dataframe:"id3.rda"
> attach(id3)
>

<snip>

>
> #create binary outcome variable (q13) (transformando q13 em binária)
> q131<-ifelse(q13==1,0,ifelse(q13==2,1,ifelse(q13==3,1,
> ifelse(q13==4,1,ifelse(q13==5,1,NA)))))

Instead of several ifelse you can use

q131 <- as.numeric(cut(id3$q13,c(0,1.5,5)))-1

> id3<-cbind(id3,q131)
> id3$q131 <- as.factor(id3$q131)
>
> str(id3)
>
> tp1 <- glm(q131 ~ q11 + q10+q12+edcat + q08+q06+ q14, family =
> binomial(link = "logit"), data=id3, na.action="na.exclude")
> tp1
>
> library(ResourceSelection)
> hoslem.test(tp1$q131, fitted(tp1), g=10)
>
>
>
>
> --------------------------------------------
> Em sex, 1/5/15, PIKAL Petr <petr.pikal at precheza.cz> escreveu:
>
>  Assunto: RE: [R] help - hoslem.test
>  Para: "Luciane Maria Pilotto" <lutipilotto at yahoo.com.br>, "r-help at r-
> project.org" <r-help at r-project.org>
>  Data: Sexta-feira, 1 de Maio de 2015, 3:59
>
>  Hi
>
>  > -----Original Message-----
>  > From: Luciane Maria Pilotto [mailto:lutipilotto at yahoo.com.br]
>  > Sent: Friday, May 01, 2015 12:49 AM
>  > To: PIKAL Petr; r-help at r-project.org;
>  John Kane
>  > Subject: Re: [R] help -
>  hoslem.test
>  >
>  > Ok, in
>  dropbox link below you can download the bank.
>  > (https://www.dropbox.com/s/9qrdf4mhd6tzypi/id3.rda?dl=0)
>
>  No. Not everybody is alowed to
>  use dropbox. Try to post result of
>
>  dput(id3)
>
>  directly to your post.
>
>  Cheers
>  Petr
>
>  >
>  > I change the script
>  following the suggestions, but the error persists.
>  > I used the the na.action command to delete
>  the lost values.
>  >
>  >
>
> #######################################################################
>  >
>  >
>  load("id3.rda")
>  >
>  attach(id3)
>  >
>  >
>  #transformando q13 em binária (o/1)
>  >
>  q131<-ifelse(q13==1,0,ifelse(q13==2,1,ifelse(q13==3,1,
>  > ifelse(q13==4,1,ifelse(q13==5,1,NA)))))
>  > id3<-cbind(id3,q131)
>  > id3$q131 <- as.factor(id3$q131)
>  > str(id3)
>  >
>  > tp1 <- glm(q131 ~ q11 + q10+q12+edcat +
>  q08+q06+ q14, family =
>  > binomial(link =
>  "logit"), data=id3,
>  na.action="na.exclude")
>  >
>  tp1
>  > hoslem.test(tp1$q131, fitted(tp1),
>  g=10)
>  >
>  >
>  >
>  >
>  --------------------------------------------
>  > Em qui, 30/4/15, John Kane <jrkrideau at inbox.com>
>  escreveu:
>  >
>  >
>  Assunto: Re: [R] help - hoslem.test
>  >
>  Para: "PIKAL Petr" <petr.pikal at precheza.cz>,
>  "Luciane Maria Pilotto"
>  >
>  <lutipilotto at yahoo.com.br>,
>  "r-help at r-project.org"
>  <r-help at r-
>  > project.org>
>  >
>  Data: Quinta-feira, 30 de Abril de 2015, 11:51
>  >
>  >  Kevin Thorpe
>  pointed out
>  >  to me that there is a
>  dropbox link at the very bottom of the
>  >  post that I missed. :(
>  >
>  >  I
>  >  just downloaded it, read it in and it
>  looks fine.
>  >
>  >  John
>  Kane
>  >  Kingston ON Canada
>  >
>  >
>  >  > -----Original
>  >  Message-----
>  >  >
>  From: petr.pikal at precheza.cz
>  >  > Sent: Thu, 30 Apr 2015 14:25:23
>  +0000
>  >  > To: lutipilotto at yahoo.com.br,
>  >  r-help at r-project.org
>  >  > Subject: Re: [R] help -
>  hoslem.test
>  >  >
>  >  > Hi
>  >  >
>  >  > I agree with John
>  >  >
>  >  > Just
>  small
>  >  refinements in lines
>  >  >
>  >  >>
>  -----Original Message-----
>  >
>  >>> -----Original Message-----
>  >  >>> From: lutipilotto at yahoo.com.br
>  >  >>> Sent: Thu, 30 Apr 2015
>  04:24:32
>  >  -0700
>  >
>  >>> To: r-help at r-project.org,
>  >  jrkrideau at inbox.com
>  >  >>> Subject: RE: [R] help -
>  >  hoslem.test
>  >
>  >>>
>  >  >>>
>  load("id3.rda")
>  >  >>
>  And what is this?
>  >  >>
>  >  >> We do not
>  >  have access to your office or computer
>  hard disc.
>  >  >>
>  >  >> Please read
>  >  http://stackoverflow.com/questions/5963269/how-to-make-a-
>  >  >> great-r-reproducible-example,
>  see
>  >  ?dput for sending data?
>  >  >>
>  >  >>
>  It is very unlikely anyone here can
>  >
>  help if we have no data.
>  >  >>
>  >  >>
>  >
>  >>>
>  >  attach(id3)
>  >  >
>  >  > Do
>  >  not use attach. It prevents from
>  modifiyng id3.
>  >  >
>  >  >>>
>  >
>  >>> #transformando q13 em binária
>  >  >>>
>  >
>  q131<-ifelse(q13==1,1,ifelse(q13==2,2,ifelse(q13==3,2,
>  >  >>>
>  >
>  ifelse(q13==4,2,ifelse(q13==5,2,NA)))))
>  >  >
>  >
>  >  > q131 <- as.numeric(cut(q13,
>  >  c(0,1.5,5)))
>  >
>  >
>  >  >>
>  >
>  x<-1:7
>  >  >> x
>  >  >
>  >  [1] 1 2 3 4
>  5 6 7
>  >  >> as.numeric(cut(x,
>  >  c(0,1.5,5)))
>  >  >
>  [1]  1  2  2  2  2 NA
>  >  NA
>  >  >
>  >
>  >>>
>  >
>  id3<-cbind(id3,q131)
>  >  >
>  >  > rather dangerous in case id3 is
>  not
>  >  data.frame but matrix
>  >  >
>  >  >>>
>  id3$q131 <-
>  >  as.factor(id3$q131)
>  >  >>>
>  >
>  >>> tp1 <- glm(q131 ~ q11 +
>  >  q10+q12+edcat + q08+q06+ q14, family
>  =
>  >  >>> binomial(link =
>  >  "logit"), data=id3)
>  >  >>>
>  >
>  tp1
>  >  >>>
>  >  >>>
>  library(ResourceSelection)
>  >
>  >>> hoslem.test(tp1$q131, fitted(tp1),
>  >  g=10)
>  >  >
>  >  > hoslem.test
>  >
>  expects x to be a numeric vector of observations, binary
>  >  > (0/1).
>  >  >
>  If I
>  >  understand correctly tp1$q131
>  have values 1, 2 or NA.
>  >  >
>  >  > Cheers
>  >  >
>  Petr
>  >  >
>  >
>  >>>
>  >  >>>
>  >  dataframe: https://www.dropbox.com/s/9qrdf4mhd6tzypi/id3.rda?dl=0
>  >  >>>
>  >
>  >>>
>  >  >>>
>  >
>  __________________________________________________
>  >  >>> Luciane Maria Pilotto
>  >  >>> Mestre e Doutoranda em
>  Saúde
>  >  Bucal Coletiva - FO/UFRGS
>  >  >>> NDE
>  >
>  Odontologia - UNIVATES
>  >
>  >>>
>  >  Telefone: (51)
>  84512344
>  >  >>>
>  >  >>>
>  >
>  --------------------------------------------
>  >  >>> Em qui, 30/4/15, John Kane
>  <jrkrideau at inbox.com>
>  >  escreveu:
>  >
>  >>>
>  >  >>>  Assunto:
>  RE: [R] help -
>  >  hoslem.test
>  >  >>>  Para:
>  >  "Luciane Maria Pilotto" <lutipilotto at yahoo.com.br>,
>  >  >>> r-help at r-project.org
>  >  >>>  Data: Quinta-feira, 30 de
>  Abril
>  >  de 2015, 7:52
>  >  >>>
>  >
>  >>>  http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-
>  >  >> reproducible-example
>  >  >>>
>  >
>  >>>
>  >  John Kane
>  >  >>>  Kingston ON
>  >  Canada
>  >
>  >>>
>  >  >>>
>  >  >>>
>  >  >
>  -----Original Message-----
>  >
>  >>>  > From: lutipilotto at yahoo.com.br
>  >  >>>  > Sent: Wed, 29 Apr
>  2015
>  >  18:45:26 -0700
>  >  >>>  > To: r-help at r-project.org
>  >  >>>  > Subject: [R] help
>  -
>  >  hoslem.test
>  >
>  >>>  >
>  >  >>>
>  > Hello,
>  >  >>>  >
>  >  >>>  > I'm working
>  with
>  >  >>>  ordinal logistic
>  regression
>  >  model (polr) and would
>  like
>  >  >>>
>  >  > to test the proportional odds
>  assumption.
>  >  >>>  For this,
>  I ran the binary
>  >  >>>  >
>  logistic
>  >  >>>  regressions
>  with varying
>  >  cutpoints on the
>  dependent
>  >  >>>
>  >  variable, as
>  >
>  >>>  > described
>  >  in the
>  following
>  >  >>>  commands.
>  >  When running the test of Hosmer and
>  >  >>>  > Lemeshow
>  (hoslem.test) for
>  >  residuals gives
>  >  >>>  error.
>  >  >>>  >
>  >  >>>  > Thanks,
>  >  >>>  > Luciane
>  >  >>>  >
>  >  >>>  >
>  >  >>>
>  >
>  ______________________________________________
>  >  >>>  > R-help at r-project.org
>  >  >>>  mailing list -- To
>  UNSUBSCRIBE
>  >  and more, see
>  >  >>>  > https://stat.ethz.ch/mailman/listinfo/r-help
>  >  >>>  > PLEASE do read the
>  posting
>  >  guide
>  >
>  >>>  > http://www.R-project.org/posting-guide.html
>  >  >>>  > and provide
>  commented,
>  >  minimal,
>  >  >>>  self-contained,
>  >  reproducible code.
>  >  >>>
>  >
>

________________________________
Tento e-mail a jakékoliv k němu připojené dokumenty jsou důvěrné a jsou určeny pouze jeho adresátům.
Jestliže jste obdržel(a) tento e-mail omylem, informujte laskavě neprodleně jeho odesílatele. Obsah tohoto emailu i s přílohami a jeho kopie vymažte ze svého systému.
Nejste-li zamýšleným adresátem tohoto emailu, nejste oprávněni tento email jakkoliv užívat, rozšiřovat, kopírovat či zveřejňovat.
Odesílatel e-mailu neodpovídá za eventuální škodu způsobenou modifikacemi či zpožděním přenosu e-mailu.

V případě, že je tento e-mail součástí obchodního jednání:
- vyhrazuje si odesílatel právo ukončit kdykoliv jednání o uzavření smlouvy, a to z jakéhokoliv důvodu i bez uvedení důvodu.
- a obsahuje-li nabídku, je adresát oprávněn nabídku bezodkladně přijmout; Odesílatel tohoto e-mailu (nabídky) vylučuje přijetí nabídky ze strany příjemce s dodatkem či odchylkou.
- trvá odesílatel na tom, že příslušná smlouva je uzavřena teprve výslovným dosažením shody na všech jejích náležitostech.
- odesílatel tohoto emailu informuje, že není oprávněn uzavírat za společnost žádné smlouvy s výjimkou případů, kdy k tomu byl písemně zmocněn nebo písemně pověřen a takové pověření nebo plná moc byly adresátovi tohoto emailu případně osobě, kterou adresát zastupuje, předloženy nebo jejich existence je adresátovi či osobě jím zastoupené známá.

This e-mail and any documents attached to it may be confidential and are intended only for its intended recipients.
If you received this e-mail by mistake, please immediately inform its sender. Delete the contents of this e-mail with all attachments and its copies from your system.
If you are not the intended recipient of this e-mail, you are not authorized to use, disseminate, copy or disclose this e-mail in any manner.
The sender of this e-mail shall not be liable for any possible damage caused by modifications of the e-mail or by delay with transfer of the email.

In case that this e-mail forms part of business dealings:
- the sender reserves the right to end negotiations about entering into a contract in any time, for any reason, and without stating any reasoning.
- if the e-mail contains an offer, the recipient is entitled to immediately accept such offer; The sender of this e-mail (offer) excludes any acceptance of the offer on the part of the recipient containing any amendment or variation.
- the sender insists on that the respective contract is concluded only upon an express mutual agreement on all its aspects.
- the sender of this e-mail informs that he/she is not authorized to enter into any contracts on behalf of the company except for cases in which he/she is expressly authorized to do so in writing, and such authorization or power of attorney is submitted to the recipient or the person represented by the recipient, or the existence of such authorization is known to the recipient of the person represented by the recipient.


More information about the R-help mailing list