search

runehaubo / lmerTestR

Repository for the R-package lmerTest
48 stars 9 forks source link

Step Function Issue #26

Closed wkmustahs21 closed 3 years ago

wkmustahs21 commented 4 years ago

Hello

I think I have an issue with output of step function. I do not know if there has been an update in the function for the change in format. I am using the function for producing 'Backward Reduced Random Effect Table' For example when I ran the code last summer the output was like this:

Model found: Shatter ~ ENTRY + REP   Backward reduced random-effect table:     Eliminated npar  logLik    AIC       LRT Df Pr(>Chisq)

                      180 -42.707 445.41 (1 \| SUB.BLOCK)          1  179 -42.707 443.41 2.558e-13  1          1 Backward reduced fixed-effect table: Eliminated  Df Sum of Sq    RSS     AIC F value Pr(>F) ENTRY          0 176    86.513 93.007 -283.75  6.9636 <2e-16 *** REP            0   1     0.006  6.500 -652.19  0.0833 0.7735 Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 |     I think the model I ran was lmer(Shatter ~ Entry + Rep + 1|Sub.Block) This time I ran with my new data but had to change the formula to: lm(Plant.Height ~Entry + Rep, data= DATH) The output now looks like this: Start: AIC=1480.97 Plant.Height ~ ENTRY_NO + REP Df Sum of Sq RSS AIC - ENTRY_NO 209 13704.6 23941 1317.9 10236 1481.0 - REP 1 71.6 10308 1481.1 Step: AIC=1317.87 Plant.Height ~ REP Df Sum of Sq RSS AIC - REP 1 70.083 24011 1316.7 23941 1317.9 Step: AIC=1316.74 Plant.Height ~ 1 ***** I could not run the the lmer() model function because I got this error: "Error in as_lmerModLmerTest(model) : model not of class 'lmerMod': cannot coerce to class 'lmerModLmerTest" Am I doing something wrong? Do I need to change the code?
runehaubo commented 4 years ago

Try to install the GitHub version of lmerTest and give it another go. If that doesn’t solve it I probably cannot help but would in any case need sessionInfo and class of the original model object to diagnose further

Hope this helps Rune

lør. 3. aug. 2019 kl. 08.17 skrev wkmustahs21 notifications@github.com:

Hello

I think I have an issue with output of step function. I do not know if there has been an update in the function for the change in format. I am using the function for producing 'Backward Reduced Random Effect Table' For example when I ran the code last summer the output was like this:

Model found: Shatter ~ ENTRY + REP

Backward reduced random-effect table:

Eliminated npar logLik AIC LRT Df Pr(>Chisq) 180 -42.707 445.41 (1 | SUB.BLOCK) 1 179 -42.707 443.41 2.558e-13 1 1

Backward reduced fixed-effect table:

     Eliminated       Df       Sum of Sq        RSS            AIC                   F value     Pr(>F)

ENTRY 0 176 86.513 93.007 -283.75 6.9636 <2e-16 *** REP 0 1 0.006 6.500 -652.19 0.0833 0.7735

Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1 |

I think the model I ran was lmer(Shatter ~ Entry + Rep + 1|Sub.Block)

This time I ran with my new data but had to change the formula to: lm(Plant.Height ~Entry + Rep, data= DATH)

The output now looks like this:

Start: AIC=1480.97 Plant.Height ~ ENTRY_NO + REP

    Df Sum of Sq   RSS    AIC
  • ENTRY_NO 209 13704.6 23941 1317.9 10236 1481.0
  • REP 1 71.6 10308 1481.1

Step: AIC=1317.87 Plant.Height ~ REP

Df Sum of Sq RSS AIC

  • REP 1 70.083 24011 1316.7 23941 1317.9

Step: AIC=1316.74 Plant.Height ~ 1

***** I could not run the the lmer() model function because I got this error:

"Error in as_lmerModLmerTest(model) : model not of class 'lmerMod': cannot coerce to class 'lmerModLmerTest"

Am I doing something wrong? Do I need to change the code?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/runehaubo/lmerTestR/issues/26?email_source=notifications&email_token=ABHZZLJYCHFW5QGW6X6SNHTQCUPIRA5CNFSM4IJCK42KYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HDFTM6A, or mute the thread https://github.com/notifications/unsubscribe-auth/ABHZZLLTNROZUJ7JOVYGMJDQCUPIRANCNFSM4IJCK42A .

wkmustahs21 commented 4 years ago

Ok I will give it a try and let you know very soon

On Sat, Aug 3, 2019 at 8:48 AM Rune Haubo B Christensen < notifications@github.com> wrote:

Try to install the GitHub version of lmerTest and give it another go. If that doesn’t solve it I probably cannot help but would in any case need sessionInfo and class of the original model object to diagnose further

Hope this helps Rune

lør. 3. aug. 2019 kl. 08.17 skrev wkmustahs21 notifications@github.com:

Hello

I think I have an issue with output of step function. I do not know if there has been an update in the function for the change in format. I am using the function for producing 'Backward Reduced Random Effect Table' For example when I ran the code last summer the output was like this:

Model found: Shatter ~ ENTRY + REP

Backward reduced random-effect table:

Eliminated npar logLik AIC LRT Df Pr(>Chisq) 180 -42.707 445.41 (1 | SUB.BLOCK) 1 179 -42.707 443.41 2.558e-13 1 1

Backward reduced fixed-effect table:

Eliminated Df Sum of Sq RSS AIC F value Pr(>F)

ENTRY 0 176 86.513 93.007 -283.75 6.9636 <2e-16 *** REP 0 1 0.006 6.500 -652.19 0.0833 0.7735

Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1 |

I think the model I ran was lmer(Shatter ~ Entry + Rep + 1|Sub.Block)

This time I ran with my new data but had to change the formula to: lm(Plant.Height ~Entry + Rep, data= DATH)

The output now looks like this:

Start: AIC=1480.97 Plant.Height ~ ENTRY_NO + REP

Df Sum of Sq RSS AIC

  • ENTRY_NO 209 13704.6 23941 1317.9 10236 1481.0
  • REP 1 71.6 10308 1481.1

Step: AIC=1317.87 Plant.Height ~ REP

Df Sum of Sq RSS AIC

  • REP 1 70.083 24011 1316.7 23941 1317.9

Step: AIC=1316.74 Plant.Height ~ 1

***** I could not run the the lmer() model function because I got this error:

"Error in as_lmerModLmerTest(model) : model not of class 'lmerMod': cannot coerce to class 'lmerModLmerTest"

Am I doing something wrong? Do I need to change the code?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub < https://github.com/runehaubo/lmerTestR/issues/26?email_source=notifications&email_token=ABHZZLJYCHFW5QGW6X6SNHTQCUPIRA5CNFSM4IJCK42KYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HDFTM6A , or mute the thread < https://github.com/notifications/unsubscribe-auth/ABHZZLLTNROZUJ7JOVYGMJDQCUPIRANCNFSM4IJCK42A

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/runehaubo/lmerTestR/issues/26?email_source=notifications&email_token=AJNDSFR3GNPXWS4YXHVYQWDQCWEB3A5CNFSM4IJCK42KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3POZCI#issuecomment-517926025, or mute the thread https://github.com/notifications/unsubscribe-auth/AJNDSFUT72QMQNPTBI2CUATQCWEB3ANCNFSM4IJCK42A .

-- Have a pleasant day! Wardah Mustahsan

wkmustahs21 commented 4 years ago

Hi Rune

I gave it another go and installed the package through devtools and github but the output is still the same and I am not sure where I am going wrong if its my model changing from lmer to lm (but I can't run step with lmer model). I will attach my session info, the output file I have, an shorter version of my data, and my R code too.

Maybe if you test it out you could catch something I missed.

Hope to hear from you soon.

Have a pleasant day! Wardah Mustahsan

On Sat, Aug 3, 2019 at 8:48 AM Rune Haubo B Christensen < notifications@github.com> wrote:

Try to install the GitHub version of lmerTest and give it another go. If that doesn’t solve it I probably cannot help but would in any case need sessionInfo and class of the original model object to diagnose further

Hope this helps Rune

lør. 3. aug. 2019 kl. 08.17 skrev wkmustahs21 notifications@github.com:

Hello

I think I have an issue with output of step function. I do not know if there has been an update in the function for the change in format. I am using the function for producing 'Backward Reduced Random Effect Table' For example when I ran the code last summer the output was like this:

Model found: Shatter ~ ENTRY + REP

Backward reduced random-effect table:

Eliminated npar logLik AIC LRT Df Pr(>Chisq) 180 -42.707 445.41 (1 | SUB.BLOCK) 1 179 -42.707 443.41 2.558e-13 1 1

Backward reduced fixed-effect table:

Eliminated Df Sum of Sq RSS AIC F value Pr(>F)

ENTRY 0 176 86.513 93.007 -283.75 6.9636 <2e-16 *** REP 0 1 0.006 6.500 -652.19 0.0833 0.7735

Signif. codes: 0 ‘’ 0.001 ‘’ 0.01 ‘’ 0.05 ‘.’ 0.1 ‘ ’ 1 |

I think the model I ran was lmer(Shatter ~ Entry + Rep + 1|Sub.Block)

This time I ran with my new data but had to change the formula to: lm(Plant.Height ~Entry + Rep, data= DATH)

The output now looks like this:

Start: AIC=1480.97 Plant.Height ~ ENTRY_NO + REP

Df Sum of Sq RSS AIC

  • ENTRY_NO 209 13704.6 23941 1317.9 10236 1481.0
  • REP 1 71.6 10308 1481.1

Step: AIC=1317.87 Plant.Height ~ REP

Df Sum of Sq RSS AIC

  • REP 1 70.083 24011 1316.7 23941 1317.9

Step: AIC=1316.74 Plant.Height ~ 1

***** I could not run the the lmer() model function because I got this error:

"Error in as_lmerModLmerTest(model) : model not of class 'lmerMod': cannot coerce to class 'lmerModLmerTest"

Am I doing something wrong? Do I need to change the code?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub < https://github.com/runehaubo/lmerTestR/issues/26?email_source=notifications&email_token=ABHZZLJYCHFW5QGW6X6SNHTQCUPIRA5CNFSM4IJCK42KYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HDFTM6A , or mute the thread < https://github.com/notifications/unsubscribe-auth/ABHZZLLTNROZUJ7JOVYGMJDQCUPIRANCNFSM4IJCK42A

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/runehaubo/lmerTestR/issues/26?email_source=notifications&email_token=AJNDSFR3GNPXWS4YXHVYQWDQCWEB3A5CNFSM4IJCK42KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3POZCI#issuecomment-517926025, or mute the thread https://github.com/notifications/unsubscribe-auth/AJNDSFUT72QMQNPTBI2CUATQCWEB3ANCNFSM4IJCK42A .

Start: AIC=1480.97 Plant.Height ~ ENTRY_NO + REP

        Df Sum of Sq   RSS    AIC

Step: AIC=1317.87 Plant.Height ~ REP

   Df Sum of Sq   RSS    AIC

Step: AIC=1316.74 Plant.Height ~ 1

Call: lm(formula = Plant.Height ~ 1, data = inputData)

Coefficients: (Intercept)
80.48

- Session info -------------------------------------------------------------------- setting value
version R version 3.5.1 (2018-07-02) os Windows >= 8 x64
system x86_64, mingw32
ui RStudio
language (EN)
collate English_United States.1252
ctype English_United States.1252
tz America/Chicago
date 2019-08-03

[1] C:/Users/wkmus/OneDrive/Documents/R/win-library/3.5 [2] C:/Program Files/R/R-3.5.1/library

[1] C:/Users/wkmus/OneDrive/Documents/R/win-library/3.5 [2] C:/Program Files/R/R-3.5.1/library Error in install.packages : Updating loaded packages

[1] C:/Users/wkmus/OneDrive/Documents/R/win-library/3.5 [2] C:/Program Files/R/R-3.5.1/library