This article presents a general approach for logit random effects modelling of
clustered ordinal and nominal responses. We review multinomial logit random effects
models in a unified form as multivariate generalized linear mixed models. Maximum
likelihood estimation utilizes adaptive Gauss-Hermite quadrature within a
quasi-Newton maximization algorithm. For cases in which this is computationally
infeasible, we generalize a Monte Carlo EM algorithm. We also generalize a
pseudo-likelihood approach that is simpler but provides poorer approximations for
the likelihood. Besides the usual normality structure for random effects, we also
present a semi-parametric approach treating the random effects in a non-parametric
manner. An example comparing reviews of movie critics uses adjacent-categories logit
models and a related baseline-category logit model.
Adams J
,
Wilson M
,
Wang W
(1997)
The multidimensional random coefficients multinomial logit model
. Applied Psychological Measurement21,
1-23
.
2.
Adams RJ
,
Wilson MR
(1996) Formulating the Rasch model as a mixed coefficients
multinomial logit. In:
Engelhard G
,
Wilson MR
, eds. Objective measurement: theory and practice. Vol. 3,
Norwood, NJ: Ablex
, 143-166.
3.
Agresti A
(1993)
Computing conditional maximum likelihood estimates for generalized Rasch
models using simple loglinear models with diagonals parameters
. Scandinavian Journal of Statistics20,
63-71
.
4.
Aitkin M
(1996)
A general maximum likelihood analysis of overdispersion in generalized
linear models
. Statistics and Computing6,
251-262
.
5.
Aitkin M
(1999)
A general maximum likelihood analysis of variance components in generalized
linear models
. Biometrics55,
117-128
.
6.
Anderson DA
,
Aitkin M
(1985)
Variance component models with binary response: interviewer variability
. Journal of Royal Statistical Society, Series B47,
203-210
.
7.
Bock RD
(1972)
Estimating item parameters and latent ability when responses are scored in
two or more nominal categories
. Psychometrika37,
29-51
.
8.
Booth JG
,
Hobert JP
(1999)
Maximizing generalized linear mixed model likelihoods with an automated
Monte Carlo EM algorithm
. Journal of the Royal Statistical Society, Series B61,
265-285
.
9.
Breslow N
(1984)
Extra-Poisson variation in log-linear models
. Applied Statistics33,
38-44
.
10.
Breslow NE
,
Clayton DG
(1993)
Approximate inference in generalized linear mixed models
. Journal of the American Statistical Association88,
9-25
.
11.
Bryk AS
,
Raudenbush SW
,
Congdon R
(2000) HLM version 5.
Chicago: Scientific Software International, Inc.
12.
Butler SM
,
Louis TA
(1992)
Random effects models with nonparametric priors
. Statistics in Medicine11,
1981-2000
.
13.
Butler SM
,
Louis TA
(1997)
Consistency of maximum likelihood estimators in general random effects
models for binary data
. The Annals of Statistics25,
351-377
.
14.
Chen Z
,
Kuo L
(2001)
A note on the estimation of the multinomial logit model with random effects
. American Statistician55,
89-95
.
15.
Chintagunta PK
,
Jain DC
,
Vilcassim NJ
(1991)
Investigating heterogeneity in brand preferences in logit models for panel data
. Journal of Marketing Research28,
417-428
.
16.
Conaway M
(1989)
Analysis of repeated categorical measurements with conditional likelihood methods
. Journal of the American Statistical Association84,
53-62
.
17.
Conaway MR
(1990)
A random effects model for binary data
. Biometrics46,
317-328
.
18.
Coull B
,
Agresti A
(2000)
Random effects modeling of multiple binary responses using the multivariate
binomial logit-normal distribution
. Biometrics56,
162-168
.
19.
Crouchley R
(1995)
A random-effects model for ordered categorical data
. Journal of the American Statistical Association90,
489-498
.
20.
Daniels MJ
,
Gatsonis C
(1997)
Hierarchical polytomous regression models with applications to health
services research
. Statistics in Medicine16,
2311-2325
.
21.
Davidian M
,
Gallant AR
(1993)
The nonlinear mixed effects model with a smooth random effects density
. Biometrika80,
475-488
.
22.
Davies RB
(1987) Mass point methods for dealing with nuisance
parameters in longitudinal studies. In:
Crouchley R
, ed. Longitudinal data analysis.
Avebury, Aldershot: Hants
, 88-109.
23.
Dempster AP
,
Laird NM
,
Rubin DA
(1977)
Maximum likelihood estimation from incomplete data via the EM algorithm
(with Discussion)
. Journal of the Royal Statistical Society, Series B39,
1-38
.
24.
Dos Santos D
,
Berridge D
(2000)
A continuation ratio random effects model for repeated ordinal responses
. Statistics in Medicine19,
3377-3388
.
25.
Engel B
(1998)
A simple illustration of the failure of PQL, IRREML and APHL as approximate
ML methods for mixed models for binary data
. Biometrical Journal40,
141-154
.
26.
Ezzet F
,
Whitehead J
(1991)
A random effects model for ordinal responses from a crossover trial
. Statistics in Medicine10,
901-907
.
27.
Fahrmeir L
,
Tutz G
(2001) Multivariate statistical modelling based on
generalized linear models. 2nd edn.New York: Springer-Verlag
.
28.
Follmann DA
,
Lambert D
(1989)
Generalizing logistic regression by nonparametric mixing
. Journal of the American Statistical Association84,
295-300
.
29.
Follmann DA
,
Lambert D
(1991)
Identifiability of finite mixtures of logistic regression models
. Journal of Statistical Planning and Inference27,
375-381
.
30.
Geweke J
(1996) Handbook of computational statistics.
New York: Elsevier
, Chapter 15.
31.
Gilmour AR
,
Anderson RD
,
Rae AL
(1985)
The analysis of data by a generalized linear mixed model
. Biometrika72,
593-599
.
32.
Goldstein H
,
Rasbash J
,
Plewis I
,
Draper D
,
Browne W
,
Wang M
(1998) A user’s guide to MLwiN. University of
London.
London: Institute of Education
.
33.
Gonul F
,
Srinivasan K
(1993)
Modeling unobserved heterogeneity in multinomial logit models:
methodological and managerial implications
. Marketing Science12,
213-229
.
34.
Greene WH
(1998) LIMDEP version 7.0 user’s manual.
Plainview, NY: Econometric Software, Inc.
35.
Hartzel, J (1999) Random effects models for nominal and ordinal
data. PhD thesis, University of Florida.
36.
Hartzel J
,
Liu I
,
Agresti A
(2001)
Describing heterogeneous effects in stratified ordinal contingency tables,
with application to multi-centre clinical trials
. Computational Statistics and Data Analysis.
37.
Harville DA
,
Mee RW
(1984)
A mixed-model procedure for analysing ordered categorical data
. Biometrics40,
393-408
.
38.
Heckman JJ
,
Singer B
(1984)
A method for minimizing the impact of distributional assumptions in
econometric models of duration
. Econometrica52,
271-320
.
39.
Hedeker D
(1999)
MIXNO: a computer program for mixed-effects nominal logistic regression
. Journal of Statistical Software4,
1-92
.
40.
Hedeker D
,
Gibbons RD
(1994)
A random-effects ordinal regression model for multilevel analysis
. Biometrics50,
933-944
.
41.
Hedeker D
,
Mermelstein RJ
(1998)
A multilevel thresholds of change model for analysis of stages of change data
. Multivariate Behavioral Research33,
427-455
.
42.
Heinen T
(1996) Latent class and discrete latent trait models:
similarities and differences.
Sage
.
43.
Hinde JP
(1982) Compound regression models. In:
Gilchrist R
, ed. GLIM 82: international conference for generalized linear
models.
New York: Springer
, 109-121.
44.
Jain DC
,
Vilcassim NJ
,
Chintagunta PK
(1994)
A random-coefficient logit brand-choice model applied to panel data
. Journal of Business and Economic Statistics12,
317-328
.
45.
Jansen J
(1990)
On the statistical analysis of ordinal data when extravariation is present
. Applied Statistics39,
74-85
.
46.
Keen A
,
Engel B
(1997)
Analysis of a mixed model for ordinal data by iterative reweighted REML
. Statistica Neerlandica51,
129-144
.
47.
Lee Y
,
Nelder JA
(1996)
Hierarchical generalized linear models (disc: P656-678)
. Journal of the Royal Statistical Society, Series B, Methodological58,
619-656
.
48.
Lin X
(1997)
Variance component testing in generalised linear models with random effects
. Biometrika84,
309-326
.
49.
Liu Q
,
Pierce DA
(1994)
A note on Gauss-hermite quadrature
. Biometrika81,
624-629
.
50.
Louis TA
(1982)
Finding the observed information matrix when using the EM algorithm
. Journal of the Royal Statistical Society, Series B44,
226-233
.
51.
McCulloch CE
(1994)
Maximum likelihood variance components estimation for binary data
. Journal of the American Statistical Association89,
330-335
.
52.
McCulloch CE
(1997)
Maximum likelihood algorithms for generalized linear mixed models
. Journal of the American Statistical Association92,
162-170
.
53.
Neuhaus JM
,
Hauck WW
,
Kalbfleisch JD
(1992)
The effects of mixture distribution misspecification when fitting
mixed-effects logistic models
. Biometrika79,
755-762
.
54.
Pinheiro JC
,
Bates DM
(1995)
Approximations to the log-likelihood function in the nonlinear
mixed-effects model
. Journal of Computational and Graphical Statistics4,
12-35
.
55.
Rasch G
(1961)
On general laws and the meaning of measurement in psychology
. In:
Neyman J
, ed. Proceedings of the 4th Berkeley symposium on mathematical
statistics and probability, vol 4. Berkeley: University of California Press.
56.
Revelt D
,
Train K
(1998).
Mixed logit with repeated choices: household choices of appliance
efficiency level
. Review of Economics and Statistics80,
647-657
.
57.
Self S
,
Liang K-Y
(1987)
Asymptotic properties of maximum likelihood estimators and likelihood ratio
tests under nonstandard conditions
. Journal of the American Statistical Association82,
605-610
.
58.
Stiratelli R
,
Laird NM
,
Ware JH
(1984)
Random effects models for serial observations with binary responses
. Biometrics40,
961-971
.
59.
Teicher H
(1963)
Indentifiability of finite mixtures
. Annals of Mathematical Statistics34,
1265-1269
.
60.
Teicher H
(1967)
Indentifiability of mixtures of product measures
. Annals of Mathematical Statistics38,
1300-1302
.
61.
Ten Have TR
(1996)
A mixed effects model for multivariate ordinal response data including
correlated discrete failure times with ordinal responses
. Biometrics52,
473-491
.
62.
Ten Have TR
,
Uttal DH
(1994)
Subject-specific and population-averaged continuation ratio logit models
for multiple discrete time survival profiles
. Applied Statistics32,
371-384
.
63.
Tutz G
,
Hennevogl W
(1996)
Random effects in ordinal regression models
. Computational Statistics and Data Analysis22,
537-557
.
64.
Wedel M
,
DeSarbo WS
(1995)
A mixture likelihood approach for generalized linear models
. Journal of Classification12,
21-55
.
65.
Wolfinger R
,
O’Connell M
(1993)
Generalized linear mixed models: a pseudo-likelihood approach
. Journal of Statistical Computation and Simulation48,
233-243
.
66.
Wood A
,
Hinde J
(1987) Binomial variance component models with a
non-parametric assumption concerning random effects. In:
Crouchley R
, ed. Longitudinal data analysis.
Avebury, Aldershot: Hants
, 110-128.
67.
Zeger SL
,
Karim MR
(1991)
Generalized linear models with random effects: a Gibbs sampling approach
. Journal of the American Statistical Association86,
79-86
.
