Longitudinal studies have known widespread use in the last years in several fields of research, as they allow to distinguish between different sources of variation. We may observe differences at the beginning of the study that stay persistent through time, and changes in the response that are due to temporal dynamics in the observed covariates. Individual-specific, time-constant, effects are often included in the linear predictor to allow for unobserved individual-specific, time constant, heterogeneity motivated by omitted individual features. The random effect approach to estimation is based on considering such effects as random variables, usually with a specific parametric distribution. This approach has been frequently criticized, as it is often employed not considering correlation between observed (i.e., covariates) and unobserved (i.e., random effects) terms. To solve this issue, we may explicitly account for correlation between observed and unobserved heterogeneity, using the so-called correlated effects approach. In this article, we show that a more general solution may be developed by estimating the random effect conditional distribution non-parametrically via a discrete probability distribution on a finite number of locations. The approach we propose is assessed via a large-scale simulation study and illustrated by the analysis of a benchmark dataset.
AbrevayaJ (2013) The projection approach for unbalanced panels. Econometrics Journal, 16, 161–78.
2.
AitkinM (1999) A general maximum likelihood analysis of variance components in generalized linear models. Biometrics, 55, 117–28.
3.
AitkinM and AlfòM (1998) Regression models for binary longitudinal responses. Statistics and Computing, 8, 289–307.
4.
AkaikeH (1973) Information theory and an extension of the maximum likelihood principle. In 2nd International Symposium on Information Theory, edited by PetrovB and CsákiF. Pages 267–81. Budapest: Akadémiai Kiadó.
5.
AlfòM and AitkinM (2006) Variance component models for longitudinal count data with baseline information: Epilepsy data revisited. Statistics and Computing, 16, 231–38.
6.
AlfòM and MaruottiA (2009) A selection model for longitudinal binary responses subject to non-ignorable attrition. Statistics in Medicine, 28, 2435–50.
7.
AndersonDA and AitkinMA (1981) Variance component models with binary response: interviewer variability. Journal of Royal Statistical Society: Series B, 47, 203–10.
8.
H-JAndreß, GolschK and SchmidtA (2013) Applied Panel Data Analysis for Economic and Social Surveys. Berlin: Springer-Verlag.
9.
AntonakisJ, BastardozN and RönkköM (2019) On ignoring the random effects assumption in multilevel models: Review, critique, and recommendations. Organizational Research Methods, 24, 443–83.
10.
BaltagiBH (2009) A Companion to Econometric Analysis of Panel Data. Chichester: John Wiley and Sons.
11.
BartolucciF, BelottiF and PeracchiF (2015) Testing for time-invariant unobserved heterogeneity in generalized linear models for panel data. Journal of Econometrics, 184, 111–23.
12.
BatesD, MächlerM, BolkerB and WalkerS (2015) Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67, 1–48.
13.
BiørnE (2001) How is generalized least squares related to within and between estimators in unbalanced panel data?Technical report. Memorandum 2001.6, University of Oslo.
14.
BöhningD (1982) Convergence of Simar’s algorithm for finding the maximum likelihood estimate of a compound Poisson process. Annals of Statistics, 10, 1006–1008.
15.
Brumback BA, DaileyAB, BrumbackLC, LivingstonMD and HeZ (2010) Adjusting for confounding by cluster using generalized linear mixed models. Statistics and Probability Letters, 80, 1650–54.
16.
BrumbackBA, ZengHW and DaileyAB (2013) Adjusting for confounding by neighborhood using generalized linear mixed models and complex survey data. Statistics in Medicine, 32, 1313–24.
17.
CeleuxG, MartinO and LavergneC (2005) Mixture of linear mixed models - application to repeated data clustering. Statistical Modelling, 5, 243–67.
18.
ChamberlainG (1980) Analysis of covariance with qualitative data. Review of Economic Studies, 47, 225–38.
19.
ChamberlainG (1982) Multivariate regression models for panel data. Journal of Econometrics, 18, 5–46.
20.
ConawayMR (1989) Analysis of repeated categorical measurements with conditional likelihood methods. Journal of the American Statistical Association, 84, 53–62.
21.
DavidianM and GallantAR (1993) The nonlinear mixed effects model with a smooth random effects density. Biometrika, 80, 475–88.
22.
DaytonCM and MacReadyGB (1988) Concomitant-variable latent-class models. Journal of the American Statistical Association, 83, 173–78.
23.
DebP and TrivediPK (1997) Demand for medical care by the elderly: a finite mixture approach. Journal of Applied Econometrics, 12, 313–36.
24.
DerSimonianR (1986) Maximum likelihood estimation of a mixing distribution. Journal of the Royal Statistical Society, Series C (Applied Statistics), 35, 302–9.
25.
DietzE and BöhningD (1994) Analysis of longitudinal data using a finite mixture model. Statistical Papers, 35, 203–10.
26.
FarberHS (1983) The determination of the union status of workers. Econometrica, 51, 1417–37.
27.
FollmannDA and LambertD (1989) Generalizing logistic regression by nonparametric mixing. Journal of the American Statistical Association, 84, 295–300.
28.
FollmannDA and LambertD (1991) Identifiability of finite mixtures of logistic regression models. Journal of Statistical Planning and Inference, 27, 375–81.
29.
FrangiL and BarisioneM (2015) Are you a union member? Determinants and trends of subjective union membership in Italian society (1972–2013). Transfer: European Review of Labour and Research, 4, 451–69.
30.
GhideyW, LesaffreE and EilersP (2005) Smooth random effects distribution in a linear mixed model. Biometrics, 60, 945–53.
31.
GoetgelukS and VansteelandtS (2008) Conditional generalized estimating equations for the analysis of clustered and longitudinal data. Biometrics, 64, 772–80.
32.
GrünB and LeischF (2008) Identifiability of finite mixtures of multinomial logit models with varying and fixed effects. Journal of Classification, 25, 225–47.
33.
GuimarãesP and PortugalP (2011) A simple feasible procedure to fit models with high-dimensional fixed effects. The Stata Journal, 14, 628–49.
34.
HahnJ and NeweyW (2004) Jackknife and analytical bias reduction for nonlinear panel models. Econometrica, 72, 1295–1319.
35.
HausmanJA (1978) Specification tests in econometrics. Econometrica, 46, 1251–71.
36.
HausmanJA and TaylorW (1981) Panel data and unobservable individual effects. Econometrica, 49, 1377–98.
37.
HeckmanJ (1981) Heterogeneity and state dependence. In Studies in Labor Markets, edited by RosenS. Chicago: University of Chicago Press.
38.
HennigC (2000) Identifiability of models for clusterwise linear regression. Journal of Classification, 17, 273–96.
39.
HindeJP and WoodATA (1987) Binomial variance component models with a non-parametric assumption concerning random effects. In Longitudinal Data Analysis, edited by CrouchleyR. Aldershot, Hants: Avebury.
40.
HsiaoC (2003) Analysis of Panel Data, 3rd edition. Cambridge: Cambridge University Press.
41.
JacobsRA, JordanMI, NowlanSJ and HintonGE (1991) Adaptive mixtures of local experts. Neural Computation, 3, 79–87.
42.
JankW and BoothJ (2003) Efficiency of Monte Carlo EM and simulated maximum likelihood in two-stage hierarchical models. Journal of Computational and Graphical Statistics, 12, 214–29.
43.
JoshiR and WooldridgeJ (2019) Correlated random effects models with endogenous explanatory variables and unbalanced panels. Annals of Economics and Statistics, 134, 243–68.
44.
KieferJ and WolfowitzJ (1956) Consistency of the maximum likelihood estimator in presence of infinitely many incidental parameters. Annals of Mathematical Statistics, 27, 886–906.
45.
KrishnakumarJ (2006) Time invariant variables and panel data models: a generalized Frisch-Waugh theorem and its implications. In Panel Data Econometrics, Volume 274: Theoretical Contributions and Empirical Applications (Contributions to Economic Analysis), edited by BaltagiB. Bingley: Emerald Group Publishing Limited.
46.
LairdNM (1978) Nonparametric maximum likelihood estimation of a mixing distribution. Journal of the American Statistical Association, 73, 805–11.
47.
LamontAE, VermuntJK and HornMLV (2016) Regression mixture models: Does modeling the covariance between independent variables and latent classes improve the results? Multivariate Behavioral Research, 51, 35–52.
48.
LeighDE (1985) The determinants of workers’ union status: Evidence from the national longitudinal surveys. The Journal of Human Resources, 20, 555–66.
49.
LesaffreE and SpiessensB (2001) On the effect of the number of quadrature points in a logistic random effects model: an example. Journal of the Royal Statistical Society, Series C, 50, 325–35.
50.
LindsayBG (1983a) The geometry of mixture likelihoods: a general theory. Annals of Statistics, 11, 86–94.
51.
LindsayBG (1983b) The geometry of mixture likelihoods, part II: the exponential family. Annals of Statistics, 11, 783–92.
52.
LitiereS, AlonsoA and MolenberghsG (2008) The impact of a misspecified random-effects distribution on the estimation and the performance of inferential procedures in generalized linear mixed models. Statistics in Medicine, 27, 3125–44.
53.
LiuQ and PierceDA (1994) A note on Gaussian-Hermite quadrature. Biometrika, 81, 624–29.
54.
LouisTA (1982) Finding the observed information matrix when using the EM algorithm. Journal of the Royal Statistical Society, B, 44, 226–33.
55.
MaddalaGS (1971) The use of variance components models in pooling cross section and time series data. Econometrica, 39, 341–58.
56.
MagderLS and ZegerSL (1996) A smooth nonparametric estimate of a mixing distribution using mixtures of Gaussians. Journal of the American Statistical Association, 91, 1141–51.
57.
MarinoMF and AlfòM (2015) Latent drop-out based transitions in linear quantile hidden Markov models for longitudinal responses with attrition. Advances in Data Analysis and Classification, 9, 483–502.
58.
MarinoMF and AlfòM (2016) Gaussian quadrature approximations in mixed hidden Markov models for longitudinal data: a simulation study. Computational Statistics and Data Analysis, 94, 193–209.
59.
MartellaF and VermuntJ (2013) Model-based approaches to synthesize microarray data: a unifying review using mixture of SEMs. Statistical Methods in Medical Research, 22, 567–82.
60.
MaruottiA and RocciR (2012) A mixed non-homogeneous hidden Markov model for categorical data, with application to alcohol consumption. Statistics in Medicine, 31, 871–86.
61.
McCullochCE (1994) Maximum likelihood estimation of variance components for binary data. Journal of the American Statistical Association, 89, 330–35.
62.
McLachlanG and PeelD (2000) Finite Mixture Models. New York: John Wiley and Sons.
63.
MundlakY (1978) On the pooling of time series and cross section data. Econometrica, 46, 69–85.
64.
MunkinMK and TrivediPK (1999) Simulated maximum likelihood estimation of multivariate mixed-poisson regression models, with application. The Econometrics Journal, 2, 29–48.
65.
NeuhausJM and KalbfleischJD (1998) Between- and within-cluster covariate effects in the analysis of clustered data. Biometrics, 54, 638–45.
66.
NeuhausJM and McCullochCE (2006) Separating between- and within- cluster covariate effects by using conditional and partitioning methods. Journal of the Royal Statistical Society, Series B, 68, 859–72.
67.
NeuhausJM and McCullochCE (2011a) Prediction of random effects in linear and generalized linear models under model misspecification. Biometrics, 67, 270–79.
68.
NeuhausJM and McCullochCE (2011b) Misspecifying the shape of a random effects distribution: Why getting it wrong may not matter. Statistical Science, 26, 388–402.
69.
S-KNg and McLachlanGJ (2005) Mixture model-based statistical pattern recognition of clustered or longitudinal data. In Proceedings of WDIC2005, edited by LovellBC and MaederA. Pages 139–44. Brisbane: APRS Workshop on Digital Image Computing.
70.
OakesD (1999) Direct calculation of the information matrix via the EM algorithm. Journal of the Royal Statistical Society, series B, 61, 479–82.
71.
PierceDA and SandsBR (1975) Extra-Bernoulli variation in binary data. Technical Report 46. Corvallis, OR: Department of Statistics, Oregon State University.
72.
PinheiroJ and BatesDM (1995) Approximations to the log-likelihood function in the nonlinear mixed-effects model. Journal of Computational and Graphical Statistics, 4, 12–35.
73.
Rabe-HesketS, SkrondalA and PicklesA (2005) Maximum likelihood estimation of limited and discrete dependent variable models with nested random effects. Journal of Econometrics, 128, 301–23.
74.
Rabe-HeskethS, SkrondalA and PicklesA (2002) Reliable estimation of generalized linear mixed models using adaptive quadrature. The Stata Journal, 2, 1–21.
75.
RizopoulosD (2012) Fast fitting of joint models for longitudinal and event time data using a pseudo-adaptive Gaussian quadrature rule. Computational Statistics and Data Analysis, 56, 491–501.
76.
RizopoulosD, MolenberghsG and VerbekeG (2008) Shared parameter models under random-effects misspecification. Biometrika, 95, 63–74.
77.
RodriguezG and EloI (2003) Intra-class correlation in random-effects models for binary data. Stata Journal, 3, 32–46.
78.
RoyJ (2003) Modelling longitudinal data with nonignorable dropouts using a latent dropout class model. Biometrics, 59, 829–36.
79.
RoyJ and DanielsM (2008) A general class of pattern mixture models for nonignorable dropout with many possible dropout times. Biometrics, 64, 538–45.
80.
RoyallRM (1986) Model robust confidence intervals using maximum likelihood estimators. International Statistical Review, 54, 221–26.
81.
SchwarzG (1978) Estimating the dimension of a model. The Annals of Statistics, 6, 461–64.
82.
SimarL (1976) Maximum likelihood estimation of a compound Poisson process. Annals of Statistics, 4, 1200–9.
83.
SkrondalA and Rabe-HesketS (2014) Handling initial conditions and endogenous covariates in dynamic/transition models for binary data with unobserved heterogeneity. Journal of the Royal Statistical Society, Series C, 63, 211–37.
84.
SongX, DavidianM and TsiatisA (2002) A semiparametric likelihood approach to joint modeling of longitudinal and time-to-event data. Biometrics, 58, 742–53.
85.
StammannA, HeissF and McFaddenD (2016) Estimating fixed effects logit models with large panel data. Technical Report. Kiel: EconStor.
86.
StataCorp (2017) Stat Longitudinal Data/Panel Data Reference Manual, Release 15. College Station, TX: Stata Press.
87.
TsiatisA and DavidianM (2004) An overview of joint modeling of longitudinal and time-to-event data. Statistica Sinica, 14, 793–818.
88.
TutzG (2025) A general framework for random effects models for binary, ordinal, count type and continuous dependent variables. Statistical Modelling.
89.
VerbeckeG and LesaffreE (1996) A linear mixed-effects model with heterogeneity in the random-effects population. Journal of the American Statistical Association, 91, 217–21.
90.
VerbeckeG and LesaffreE (1997) The effect of misspecifying the random-effects distribution in linear mixed models for longitudinal data. Computational Statistics and Data Analysis, 23, 541–56.
91.
VermuntJK (2007) A hierarchical mixture model for clustering three-way data sets. Computational Statistics and Data Analysis, 51, 5368–76.
92.
VermuntJK (2010) Latent class modeling with covariates: Two improved three-step approaches. Political Analysis, 18, 450–69.
93.
ViroliC (2011) Finite mixtures of matrix normal distributions for classifying three-way data. Statistics and Computing, 21, 511–22.
94.
VivianiS, AlfòM and BruttiP (2013) Sparse nonparametric graphical models for random effect distribution in GLMMs. In 47th Meeting of the Italian Statistical Society, Advances in Latent Variables-Methods, Models and Applications. Rome: Italian Statistical Society
95.
WangP, PutermanML, CockburnI and LeN (1996) Mixed Poisson regression models with covariate dependent rates. Biomeetrics, 52, 381–400.
96.
WangY and TaylorJ (2001) Jointly modeling longitudinal and event time data with application to acquired immunodeficiency syndrome. Journal of the American Statistical Association, 96, 895–905.
97.
WedelM and DeSarboWS (1995) A mixture likelihood approach for generalized linear models. Journal of Classification, 12, 21–55.
98.
WooldridgeJM (2005) Simple solutions to the initial conditions problem in dynamic, nonlinear panel data models with unobserved heterogeneity. Journal of Applied Econometrics, 20, 39–54.
99.
WooldridgeJM (2009) Correlated random effects models with unbalanced panels. Working Paper. East Lansing, MI: Department of Economics, Michigan State University.
100.
WooldridgeJ (2019) Correlated random effects models with unbalanced panels. Journal of Econometrics, 211, 137–50.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
