Questions in topic: "mixedmodels"
https://developer.ibm.com/answers/questions/topics/single/403825.html
The latest questions for the topic "mixedmodels"Linear Mixed Model repeated covariance type
https://developer.ibm.com/answers/questions/512929/linear-mixed-model-repeated-covariance-type.html
Hello everyone :)
I need your help with regards to specifying repeated covariance type in SPSS.
I am trying to build a Linear Mixed Model in SPSS with the subject being 'borough' and repeated variable being YEAR . My dependent variable is the number of people claiming housing benefit, per year (for 7 years overall), for each borough. All my covariates (fixed effects) represent a value for a tax year. Examples of my covariates are: population, unemployment rate, mean rent, median house price to median earnings ratio. I've tried setting different covariance types and the results do differ significantly, so I guess it is important to set the correct one. I've tried researching, how to determine the repeated covariance type but I've been unsuccessful. This is the only guidance I've found https://www.ibm.com/support/knowled...ariance_structures.html#covariance_structures however, as I am not a statistician (but a sociologist) these descriptions mean nothing to me and sound like a foreign language.
The covariance types options are:
- Ante-Dependence:First Order
- AR1
- AR1 Heterogenous
- ARMA
- Compound Symmetry
- Compound Symmetry: Correlation Metric
- Compound Symmetry: Heterogenous
- Diagonal Factor Analytic: First Order
- Factor Analytic: First Order Heterogeneous
- Huynh-Feldt
- Scaled Identity
- Toeplitz
- Toeplitz: Heterogenous
- Unstructured
- Unstructured:Correlations
I would be grateful for your help!linearmixedmodelsSat, 27 Jul 2019 16:57:22 GMTmaggieldnLinear Mixed models, Warning
https://developer.ibm.com/answers/questions/506048/linear-mixed-models-warning.html
Hello,
I get the following warning when I try to run a mixed model:
"The levels of the repeated effect are not different for each observation within a repeated subject, This command is not executed
What goes wrong? How can I solve this?
My data looks as follow:
Subject - Product - Time - Session - DV (not shown below)
1111
1121
1131
1141
1151
1161
1112
1122
1132
1142
1152
1162
1211
1221
1231
1241
1251
1261
1212
1222
1232
1242
1252
1262
etc.
Thank you in advance!warningmixedmodelsMon, 27 May 2019 13:56:41 GMTMAR0902Linear Mixed-Effects Model error message
https://developer.ibm.com/answers/questions/505775/linear-mixed-effects-model-error-message.html
There are two fixed factors in my LMM: the between-subjects factor of Group and the within-subjects factor of Task. There is the single random factor of Participant nested within Group.
I am getting the **"levels of the repeated effect are not different for each observation within a repeated subject"** error message when trying to run my analysis. I put the Participant factor in the Subjects box and the Task factor in the Repeated box on the first prompt.
I have looked at lots of information on this, including this from IBM - https://www-01.ibm.com/support/docview.wss?uid=swg21479744 - but I have not been able to resolve the problem. The info I've found is geared towards situations when there are single measurements per repeated condition. The issue seems to be that I have **raw data** (RTs) and, thus, multiple measurements per repeated condition - each participant has about 50 trials per Task. Thus, a typical participant looks something like this:
Part. Group Task RT
1 1 1 .015
1 1 1 .024
1 1 1 .073
1 1 1 .009
and so on.
Any suggestions? Do I need to transform the data itself, insert some syntax, or what? How significant are the ramifications if I treat the repeated factor as a between-groups factor?linearmixedmodelsThu, 23 May 2019 15:04:50 GMTPaul SanfordHow to do post-hoc tests in SPSS MIXED procedure
https://developer.ibm.com/answers/questions/481279/how-to-do-post-hoc-tests-in-spss-mixed-procedure.html
I'm using MIXED procedure to look at group by time interactions. The output produced by the EM Means tab provides allows me to see mean differences by group at each time point, however it does not provide a p-value. Why would this be?ecmsupportpairwise-comparisonsmixedmodelsinteractionposthocTue, 20 Nov 2018 13:57:08 GMTSANJAKGLMM running continuously without producing any output
https://developer.ibm.com/answers/questions/473282/glmm-running-continuously-without-producing-any-ou.html
I've been trying to run a GLMM analysis for hours now but SPSS does not produce results. To make sure that this problem was not caused by my own wrongdoing I tried running the [IBM tutorial GLMM analysis][1] (using the [cable_survey.sav][2] data provided by IBM). However, even with the latter data set and following the instructions from the tutorial carefully, I encounter the same problem: GLMM runs for over an hour without producing output. I use **SPSS 25** on a MacBook Pro. Below is the syntax that I used, please help!
DATASET ACTIVATE DataSet1.
*Generalized Linear Mixed Models.
GENLINMIXED
/DATA_STRUCTURE SUBJECTS=customer_id
/FIELDS TARGET=service_usage TRIALS=NONE OFFSET=NONE
/TARGET_OPTIONS DISTRIBUTION=MULTINOMIAL LINK=LOGIT
/FIXED EFFECTS=edcat inccat reside service_type USE_INTERCEPT=TRUE
/RANDOM EFFECTS=customer_id USE_INTERCEPT=FALSE SUBJECTS=customer_id
COVARIANCE_TYPE=VARIANCE_COMPONENTS SOLUTION=FALSE
/BUILD_OPTIONS TARGET_CATEGORY_ORDER=ASCENDING INPUTS_CATEGORY_ORDER=ASCENDING MAX_ITERATIONS=100
CONFIDENCE_LEVEL=95 DF_METHOD=RESIDUAL COVB=MODEL PCONVERGE=0.000001(ABSOLUTE) SCORING=0
SINGULAR=0.000000000001
/EMMEANS_OPTIONS SCALE=ORIGINAL PADJUST=LSD.
[1]: https://www.ibm.com/support/knowledgecenter/en/SSLVMB_24.0.0/components/glmm/glmm_cablesurvey_howto.html
[2]: https://www3.nd.edu/~kkelley/Teaching/Data/cable_survey.savmixedmodelsWed, 03 Oct 2018 15:12:17 GMTTengelsI am using SPSS Linear Mixed Models and this error appears: "model cannot be fitted because number of observations is less than or equal to number of model parameters". What does this mean and how do I fix it?
https://developer.ibm.com/answers/questions/471851/i-am-using-spss-linear-mixed-models-and-this-error.html
I am using SPSS Linear Mixed Models and this error appears: "model cannot be fitted because number of observations is less than or equal to number of model parameters". What does this mean and how do I fix it?spssstatisticsmixedmodelsMon, 24 Sep 2018 21:22:02 GMTjgrumsHesse Matrix is not positively definite - empty multilevel model
https://developer.ibm.com/answers/questions/461953/hesse-matrix-is-not-positively-definite-empty-mult.html
I am currently working on a multilevel model on the individual and small area effect on health which is why I am conducting a linear multilevel analysis with SPSS 24. When modelling the data in an empty model (only including outcome (birth weight) and the group identifier variable (about 80 level 2 units) , the following error occured:
"The final Hessian matrix is not positive definite although all convergence criteria are satisfied. The MIXED procedure continues despite this warning. Validity of subsequent results cannot be ascertained."
SPSS delivers all the required outputs, but the covariance and standard error for the random intercept are zero.
The problem seems to be a filter variable that is included which eliminates all multiple (n=21) and preterm births (n=50). If the multiple births are excluded, the error message occurs. And I can't imagine why, because the number of excluded cases is very low and the variance in the outcome is still high. If only preterm births are excluded, no error message occurs. Including these two variables in the model as covariates seems unwise from a content-related point of view, as these would explain the biggest part of the model. When I continue to include more variables in the model, the error message disappears; the covariance parameter for the random intercept is not significant.
I have tried all the tips I found here in the forum and many others such as: another covariance structure (VC, Identity, UN), increasing the number of iterations, standardizing the variables, trying other models (GENLIN mixed - logistic, Poisson, linear) but none worked out.
Is there anything else I could try? Or is a multilevel model not the right choice for my data?errorfiltermatrixmixedmodelsTue, 31 Jul 2018 19:48:03 GMTLisaWHow do I get the results of estimated mean changes from mixed model?
https://developer.ibm.com/answers/questions/459342/how-do-i-get-the-results-of-estimated-mean-changes.html
Is there any way to get the results of estimated mean changes in LMM in SPSS?
For instance, there are 3 "groups" and repeated measures of "scores" by time, how can I get the results of estimated mean changes of scores of each group with SE? (I mean the changes of the scores of each group by time (pre-post))
Please, help me. thanks.mixedmodelsThu, 19 Jul 2018 05:11:27 GMTchoich99999How do I plot/graph estimated means from mixed model?
https://developer.ibm.com/answers/questions/453023/how-do-i-plotgraph-estimated-means-from-mixed-mode.html
How do I plot/graph estimated means from mixed model analysis?mixedmodelsTue, 12 Jun 2018 21:05:08 GMTMonhaBootstrapping and mixed model - how to write a syntax command
https://developer.ibm.com/answers/questions/445711/bootstrapping-and-mixed-model-how-to-write-a-synta.html
My question is regarding bootstrapping and the linear mixed model for longitudinal data. How could a bootstrapping syntax command be built up to fit a mixed model procedure?statisticsbootstrapmixedmodelsWed, 02 May 2018 12:30:00 GMTAnaPlaMIXED Command: Is it possible to constrain error variance to zero?
https://developer.ibm.com/answers/questions/432060/mixed-command-is-it-possible-to-constrain-error-va.html
I am analyzing dyadic data using MLM with SPSS. I'm following the procedures outlined in Dyadic Data Analysis by Kenny, Kashy & Cook.
Based on the book, I should be using the two-intercept model which requires that error variance in constrained to zero. When this book was published in 2006, SPSS was not able to fix the error variance to zero. Does anyone know if SPSS is capable of doing this now?
The syntax I am using is:
MIXED Satisfaction_A BY Role WITH Egender EAge EExperience Gender Age Dummy_Customer Dummy_Employee
EmoSupport_A InstrumSupport_A EmoSupport_P InstrumSupport_P
/FIXED=Egender EAge EExperience Gender Age Dummy_Customer Dummy_Employee
Dummy_Customer*EmoSupport_A Dummy_Customer*InstrumSupport_A Dummy_Customer*EmoSupport_P
Dummy_Customer*InstrumSupport_P Dummy_Employee*EmoSupport_A Dummy_Employee*InstrumSupport_A
Dummy_Employee*EmoSupport_P Dummy_Employee*InstrumSupport_P | NOINT SSTYPE(3)
/METHOD=ML
/PRINT=SOLUTION TESTCOV
/RANDOM=Dummy_Customer Dummy_Employee | SUBJECT(FocalID) COVTYPE(UNR)
/REPEATED=Role | SUBJECT(Dyad_ID*FocalID) COVTYPE(UNR).
In order to fix the error variance to zero in SAS, the following syntax would be added:
PARMS 2, 1, 2, 0.000001 / HOLD=4
So is there a comparable command I can use in SPSS?
Thanks!spssstatisticssyntaxmixedmodelsSun, 18 Feb 2018 11:38:49 GMTjpsiegelMIxed model takes too long in SPSS
https://developer.ibm.com/answers/questions/425233/mixed-model-takes-too-long-in-spss.html
As part of my statistics learning, I am trying to replicate results from studies that were analyzed by an experienced statisticians. In one study, they used mixed linear model to analyze the effect of an intervention on a continuous dependent variable (blood pressure) over 20 visits to the clinic.The number of observations is around 3500 blood pressure readings. They stated in the analysis plan that they used unstructured covariance matrix. When I try to run the model with the same specifications, I don't get results even after 1 hour and the status bar is showing that SPSS engine is running. Am I doing something wrong that causes the program to hang? and is there any way to speed up the processing time?mixedmodelsThu, 18 Jan 2018 06:03:59 GMTMando_87Can you calculate an effect size for slopes in mixed models?
https://developer.ibm.com/answers/questions/425074/can-you-calculate-an-effect-size-for-slopes-in-mix.html
I have been working on a multilevel project using simple slopes analysis through mixed models in SPSS.
I originally reported the intercepts and slopes for my relationships along with their significance tests. A colleague has mentioned that an effect size for the relationships (i.e. the slopes) would supplement my results but I have not been successful in finding a way to calculate this.
Does anyone have any suggestions?spssstatisticsmixedmodelsWed, 17 Jan 2018 15:14:38 GMTKDoyle32Mixed-effect ordinal regression
https://developer.ibm.com/answers/questions/414075/mixed-effect-ordinal-regression.html
I would like to analyse drug effects (drug A / B vs control) on the development of disease (ordinal range of severeness 0-5) in an animal model.
Does anybody know how to build up my syntax in a matter that I can compare my groups over time (20+ time points)?mixedmodelsordinalregressionMon, 20 Nov 2017 12:52:07 GMTDBroekaartRepeated measures MIXED ANOVA Error
https://developer.ibm.com/answers/questions/413363/repeated-measures-mixed-anova-error.html
Hello,
I am having a horrible time trying to analyse my data.
After reading a lot I decided that I need to use a RM Mixed ANOVA.
I am trying to find out if production of PGF2a varies between control and three concentrations of LPS and across time (24, 48 and 72h). I have data from 6 different horses.
My data looks like this:
![alt text][1]
where 1 is control, 2 is lps low concentration, 3 is lps mid concentration and 4 is lps high concentration.
each horse has a different number to differentiate them
I did: Analyze - Mixed Models - Linear
Horse as subject, Timepoint as Repeated
Compound symmetry for Repeated Covariance Type
PGF2aProduction as dependent variable
Treatment and timepoint as factors
Horse as Random effect
and I get an error:
MIXED LOG10_PGF2a_production BY Treatment Timepoint
/CRITERIA=CIN(95) MXITER(100) MXSTEP(10) SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0,
ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE)
/FIXED=| SSTYPE(3)
/METHOD=REML
/PRINT=SOLUTION TESTCOV
/REPEATED=Timepoint | SUBJECT(Horse) COVTYPE(CS)
/SAVE=FIXPRED PRED RESID
/EMMEANS=TABLES(OVERALL).
Warnings
|The levels of the repeated effect are not different for each observation within a repeated subject.|
|Execution of this command stops.
Can someone help please?
Many thanks!
[1]: /answers/storage/temp/18595-untitled.jpgspssanovamixedmodelsFri, 17 Nov 2017 07:18:04 GMTMaitheKeep getting error messages for Mixed Effect Model - Urgent help needed!!
https://developer.ibm.com/answers/questions/413693/keep-getting-error-messages-for-mixed-effect-model-1.html
Can anyone offer me some advice on how to organise my data for Mixed Effect Model in SPSS??? I keep getting error messages:' The levels of the repeated effect are not different for each observation within a repeated subject.'
I have two groups of participants X and Y who performed two sets of tasks i.e. A and B. Group 1 has 12 subjects, Group 2 has 16 subjects. Each subjects performed 12 trials in each of the tasks although there are some missing data. I'm interested in the differences in latency (Dependent Variable) between the two groups and tasks and would like to use Linear Mixed Model to analyse the data in SPSS. I labelled Task as Repeated Measures factor (same group of participants in both tasks), Group as Fixed Effect, and ID as Random Effect.
I've organised my data in this manner:
ID Group Task Latency
1 A 1 283
1 A 1 357
1 A 1 457
2 A 1 289
2 A 1 354
2 A 1 892
3 B 1 283
3 B 1 357
3 B 1 457
3 B 1 289
4 B 1 354
4 B 1 892
4 B 1 283
1 A 2 283
1 A 2 357
1 A 2 457
2 A 2 289
2 A 2 354
2 A 2 892
3 B 2 283
3 B 2 357
3 B 2 457
3 B 2 289
4 B 2 354
4 B 2 892
4 B 2 283spssstatisticserrordatamixedmodelsThu, 16 Nov 2017 19:45:39 GMTStacyFKeep getting error messages for Mixed Effect Model - Urgent help needed!!
https://developer.ibm.com/answers/questions/413687/keep-getting-error-messages-for-mixed-effect-model.html
Can anyone offer me some advice on how to organise my data for Mixed Effect Model in SPSS??? I keep getting error messages:' The levels of the repeated effect are not different for each observation within a repeated subject.'
I have two groups of participants X and Y who performed two sets of tasks i.e. A and B. Group 1 has 12 subjects, Group 2 has 16 subjects. Each subjects performed 12 trials in each of the tasks although there are some missing data. I'm interested in the differences in latency (Dependent Variable) between the two groups and tasks and would like to use Linear Mixed Model to analyse the data in SPSS. I labelled Task as Repeated Measures factor (same group of participants in both tasks), Group as Fixed Effect, and ID as Random Effect.
I've organised my data in this manner:
ID Group Task Latency
1 A 1 283
1 A 1 357
1 A 1 457
2 A 1 289
2 A 1 354
2 A 1 892
3 B 1 283
3 B 1 357
3 B 1 457
3 B 1 289
4 B 1 354
4 B 1 892
4 B 1 283
1 A 2 283
1 A 2 357
1 A 2 457
2 A 2 289
2 A 2 354
2 A 2 892
3 B 2 283
3 B 2 357
3 B 2 457
3 B 2 289
4 B 2 354
4 B 2 892
4 B 2 283statisticsmixedmodelsThu, 16 Nov 2017 19:35:26 GMTStacyFI need to carry out a linear mixed models with post hoc Sidek. Is this possible with version 25?
https://developer.ibm.com/answers/questions/412110/i-need-to-carry-out-a-linear-mixed-models-with-pos.html
I need to carry out a linear mixed models with post hoc Sidek. I can't seem to figure out how to do this on version 25. Any help is appreciated. I am new to SPSS!linearmixedmodelsposthocThu, 09 Nov 2017 20:52:27 GMTcdezynskiHelp translating SAS code for Mixed Models with repeated measures
https://developer.ibm.com/answers/questions/403637/help-translating-sas-code-for-mixed-models-with-re.html
I'm looking for the SPSS syntax or point and click settings that would run the same as model as this SAS code:
PROC MIXED DATA=datafile ORDER=internal;
CLASS gender group time id;
MODEL ptsd= time|group|gender;
REPEATED time / SUBJECT=id GROUP=group TYPE=un;
Thank you!spssstatisticsspssstudentmixedmodelsWed, 27 Sep 2017 14:42:53 GMTAlMac