### Citations

4643 | GG: The measurement of observer agreement for categorical data - JR, Koch - 1977 |

3157 |
Generalized Linear Models
- McCullagh, Nelder
- 1989
(Show Context)
Citation Context ...nction. Moreover, the transformation of the expected value of the response variable that yields the linear predictor that can be restricted by a regression model, is referred to as the link function (=-=McCullagh and Nelder, 1983-=-). Below, we define the linear predictors and the corresponding regression models for categorical, count, and continuous response variables. 2.2.1 Nominal and ordinal dependent variables Let us first ... |

2757 | Statistical analysis with missing data
- Little, Rubin
- 1987
(Show Context)
Citation Context ...using all available information for each of the cases. The assumption that is made is that the missing data are missing at random (MAR) or, equivalently, that the missing data mechanism is ignorable (=-=Little and Rubin, 1987-=-; Schafer, 1997; Skrondal and Rabe-Hesketh, 2004; Vermunt, 1997). Although conceptually similar, technically things are somewhat more complicated with multivariate densities for multiple response vari... |

2186 | Bayesian data analysis - Gelman, Carlin, et al. - 2003 |

1980 |
Practical optimization
- Gill, Murray, et al.
- 1981
(Show Context)
Citation Context ...ted predictor effects, and positive error variances and dispersion parameters – are dealt with using an active-set variant of the Newton-Raphson method described above (Galindo, Vermunt, Croon, 2001; =-=Gill, Murray, and Wright, 1981-=-). For that purpose, the effects involved in the order constraints are reparameterized so that they can be impose using simple nonnegativity constraints of the form ϑ ≥ 0. In an active-set method, the... |

1414 |
Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks
- Long
- 1997
(Show Context)
Citation Context ...egression Module, it is also possible to specify models for overdispersed Poisson counts. By assuming that the Poisson rate follows a gamma distribution, one obtains a negative-binomial distribution (=-=Long, 1997-=-; Simonoff, 2003). The exact form of this distribution is P (yit|x, zi, eit) = Γ(yit + 1/σ 2 t,x) yit! Γ(1/σ2t,x) ( 1/σ2t,x 1/σ2t,x + µt,x,i )1/σ2t,x ( µt,x,i 1/σ2t,x + µt,x,i )yit , where Γ(·) is the... |

787 |
Multilevel statistical models
- Goldstein
- 1995
(Show Context)
Citation Context ...Another important application of CFactors involves random-effects modeling using any of the generalized linear models implemented in the Regression 85 Module (Agresti, Booth, Hobert, and Caffo, 2000; =-=Goldstein, 1995-=-; Hedeker, 2003; Hedeker and Gibbons, 1996; Snijders and Bosker, 1999; Skrondal and Rabe-Hesketh, 2004; McFadden and Train, 2000; Wong and Mason, 1985).42 Let us first look at the random intercept mod... |

627 |
Mixture Models: Inference and Application to Clustering
- McLachlan, Basford
- 1988
(Show Context)
Citation Context ...mbership probabilities to classify cases into the appropriate cluster. The most popular model-based approach is known as mixture-model clustering, where each latent class represents a hidden cluster (=-=McLachlan and Basford, 1988-=-, Vermunt and Magidson, 2002a). Within the marketing research field, this method is sometimes referred to as “latent discriminant analysis” (Dillon and Mulani, 1999). Today's high-speed computers make... |

606 |
Maximum likelihood estimation from incomplete data via the em algorithm (with discussion
- Dempster, Laird, et al.
- 1977
(Show Context)
Citation Context ... (x|zi,ϑ)f(yi|x, zi,ϑ) ∂ϑ , (14) where wxi = wi P (x|zi,yi,ϑ) = wi P (x|zi,ϑ)f(yi|x, zi,ϑ) f(yi|zi, ϑ) . (15) 49 The EM algorithm is a general method for dealing with ML estimation with missing data (=-=Dempster, Laird, and Rubin, 1977-=-; McLachlan and Krishnan, 1997). This method exploits the fact that the first derivatives of the incomplete data log-likelihood (logL) equal the first derivatives of the complete data log-likelihood (... |

469 |
Model-Based Gaussian and NonGaussian Clustering
- Banfield, Raftery
- 1993
(Show Context)
Citation Context ... come from class-specific multivariate normal distributions f(yi|x) = (2pi)−Km/2 |Σx|−1/2 exp { −1 2 (yi − µx)′Σ−1x (yi − µx) } This model is also known as an unrestricted FM of multivariate normals (=-=Banfield and Raftery, 1993-=-; McLachlan and Basford, 1988; McLachlan and Peel, 2000; Wolfe, 1970). As can be seen, each latent class (or mixture component) has its own set of means µx and its own variance-covariance matrix Σx. S... |

455 |
Latent Variable Models and Factor Analysis
- Bartholomew, Knott
- 1999
(Show Context)
Citation Context ...ndicators (λtx1 = λx1). 9.2.2 IRT models The main difference between factor analysis and IRT modeling is that in the former response variables are continuous, whereas in the latter they are discrete (=-=Bartholomew and Knott, 1999-=-). With the Latent GOLD CFactors option, it is possible to obtain (marginal) ML estimates for the most important types of parametric IRT models for binary, ordinal, and nominal response variables, as ... |

376 |
Some latent trait models and their use in inferring an examinee's ability
- Birnbaum
- 1968
(Show Context)
Citation Context ...d Swaminathan (1985), Heinen (1996), Mellenbergh (1995), and Van der Linden and Hamilton (1997). 83 for dichotomous, ordinal, and nominal responses, respectively, yielding the two-parameter logistic (=-=Birnbaum, 1968-=-), the generalized partial-credit (Muraki, 1992), and the nominal-response model (Bock, 1972). It should be noted that the parameterization used in IRT modeling is sometimes a bit different from the o... |

364 | Latent structure analysis - Lazarsfeld, Henry - 1968 |

346 | Finite Mixtures Distributions - Everitt, Hand - 1981 |

323 |
A rating formulation for ordered response categories,”
- Andrich
- 1978
(Show Context)
Citation Context ...edure will always imply certain restrictions on the intercept across replications. These restrictions turn out to yield a restricted variant of the partial-credit model called the rating-scale model (=-=Andrich, 1978-=-), in which βtm0 = β . m0+ y ∗t m · βt.0. 41It is also possible to assume that a = 1 and treat the variance of the latent variable as an unknown parameter to be estimated (see also discussion on rando... |

323 | Item response theory: Principles and applications.Boston - Hambleton, Swaminathan - 1985 |

319 | Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm - Bock, Aitkin - 1981 |

256 |
Exploratory latent structure analysis using both identifiable and unidentifiable models
- Goodman
- 1974
(Show Context)
Citation Context ...en class membership. This is equivalent to the assumption of local independence that is common in latent variable models, including in the traditional latent class model (Bartholomew and Knott, 1999; =-=Goodman, 1974-=-a, 1974b; Magidson and Vermunt, 2004). Also in random-coefficients models, it is common to assume responses to be independent conditional on the value of the random coefficients. 2.2 Rankings and Othe... |

237 | Illustration of Bayesian inference in Normal data models using Gibbs sampling - Gelfand, Hills, et al. - 1990 |

215 | Latent class analysis - McCutcheon - 1987 |

184 |
Estimating item parameters and latent ability when responses are scored in two or more nominal categories.
- Bock
- 1972
(Show Context)
Citation Context ... 83 for dichotomous, ordinal, and nominal responses, respectively, yielding the two-parameter logistic (Birnbaum, 1968), the generalized partial-credit (Muraki, 1992), and the nominal-response model (=-=Bock, 1972-=-). It should be noted that the parameterization used in IRT modeling is sometimes a bit different from the one used here. For example, the two-parameter logistic model is usually written as ηtFi = a t... |

179 |
An introduction to categorical data analysis. Second Edition.
- Agresti
- 2007
(Show Context)
Citation Context ...ew tool to identify important market segments in target marketing. Recently, the close connection between LC models and random effects models (REM) has been made apparent (Vermunt and Van Dijk, 2002; =-=Agresti, 2002-=-). In addition, the close connection between latent classes and “hidden layer nodes” in the most widely used neural net model, the multilayer perceptron (MLP) has been clarified (Vermunt and Magidson,... |

143 |
Nonparametric Maximum Likelihood Estimation of a Mixing
- Laird
- 1978
(Show Context)
Citation Context ...atives when expressing that segment’s preferences. Such a discrete characterization of unobserved heterogeneity is sometimes referred to as a nonparametric random-coefficients approach (Aitkin, 1999; =-=Laird, 1978-=-; Vermunt, 1997; Vermunt and Van Dijk, 2001; Vermunt and Magidson, 2003). Latent GOLD Choice implements a nonparametric variant of the random-coefficient or mixed conditional logit model (Andrews et a... |

132 | Train (2000) "Mixed MNL Models for Discrete Response - McFadden, K |

118 |
The Choice Theory Approach to Marketing Research
- McFadden
- 1986
(Show Context)
Citation Context ... models, and more recently with discrete choice and ranking models. Discrete choice modeling allows estimation of consumer utilities, a theoretical development that led to the award of a Nobel prize (=-=McFadden, 1986-=-). LC choice models provide an important advance in these models, allowing for different utilities to be estimated for each latent segment (McFadden and Train, 2000, Greene and Hensher, 2002, Vermunt ... |

103 | A latent class model for discrete choice analysis: contrasts with mixed logit.
- Greene, Hensher
- 2003
(Show Context)
Citation Context ...d of a Nobel prize (McFadden, 1986). LC choice models provide an important advance in these models, allowing for different utilities to be estimated for each latent segment (McFadden and Train, 2000, =-=Greene and Hensher, 2002-=-, Vermunt and Magidson, 2003). Like traditional regression modeling, LC regression requires a computer program. As LC regression modeling is relatively new, very few programs currently exist. Our comp... |

95 |
Simple models for the analysis of association in cross-classifications having ordered categories.
- Goodman
- 1979
(Show Context)
Citation Context ...now compared with the average (geometric mean) of the probabilities of all M categories. 14 With an ordinal dependent variable we make use of the adjacent-category ordinal logit model (Agresti, 2002; =-=Goodman, 1979-=-; Magidson, 1996) in which ηm|zi = βm0 + P∑ p=1 β·p · y∗m · zip, irrespective of whether one uses dummy or effect coding for the dependent variable. As can be seen, compared to the nominal logit model... |

84 |
Finite-mixture structural equation models for response-based segmentation and unobserved heterogeneity.
- Jedidi, Jagpal, et al.
- 1997
(Show Context)
Citation Context ... t m0 + λ t . · y∗tm · Fi, ηtm,Fi = β t m0 + λ t m · Fi, 39This is the most important special case of the more general framework of mixture structural equation modeling (Dolan and Van der Maas, 1997; =-=Jedidi, Jagpal, and DeSarbo, 1997-=-). 40Overviews of the field of parametric IRT modeling can be found in Hambleton and Swaminathan (1985), Heinen (1996), Mellenbergh (1995), and Van der Linden and Hamilton (1997). 83 for dichotomous, ... |

83 |
Latent class and discrete latent trait models: Similarities and differences. Thousand Oaks:
- Heinen
- 1996
(Show Context)
Citation Context ...ful in scaling and IRT-like applications. With dichotomous indicators, for example, it yields a LC or nonparametric Rasch model and with ordinal indicators a LC or nonparametric partial-credit model (=-=Heinen, 1996-=-; Vermunt, 2001). • Order-restricted Clusters. With this constraint, cluster-specific item probabilities and means will be restricted to be monotonically increasing. It yields what is usually referred... |

75 | Estimating outcome distributions for compliers in instrumental variables models”, - Imbens, Rubin - 1997 |

74 | Latent class models - Clogg - 1995 |

70 | Bayesian Estimation of a Multilevel IRT Model using Gibbs Sampling. Psychometrika 66:271–188.
- Fox, Glas
- 2001
(Show Context)
Citation Context ...n be modelled using continuous random effects (GCFactors) and/or discrete random effects (GClasses). 10.2.5 Various IRT applications Several authors have proposed multilevel extensions of IRT models (=-=Fox and Glas, 2001-=-; Skrondal and Rabe-Hesketh, 2004). A possible specification of a multilevel IRT model is ηtFji,F gj = β t 0 + λ t · Fji + λt,g · F gj , which besides the lower-level trait Fji contains a higher-level... |

69 |
Categorical Longitudinal Data–Loglinear Analysis of Panel, Trend and Cohort Data.
- Hagenaars
- 1990
(Show Context)
Citation Context ...or response variable yit. 3.3 Covariates An important extension of the standard LC model described above is obtained by the possibility of including covariates (Clogg, 1981; Dayton and McReady, 1988; =-=Hagenaars, 1990-=-, 1993).9 Inclusion of covariates to predict class membership is straightforward within the general framework of the model defined in equation (1). Suppose we have a model with three categorical indic... |

64 | Concomitant-Variable Latent-Class Models," - Dayton, MacReady - 1988 |

63 |
The Analysis of Systems of Qualitative Variables When Some of the Variables Are Unobservable. Part I-A Modified Latent Structure Approach,"
- Goodman
- 1974
(Show Context)
Citation Context ...en class membership. This is equivalent to the assumption of local independence that is common in latent variable models, including in the traditional latent class model (Bartholomew and Knott, 1999; =-=Goodman, 1974-=-a, 1974b; Magidson and Vermunt, 2004). Also in random-coefficients models, it is common to assume responses to be independent conditional on the value of the random coefficients. 2.2 Rankings and Othe... |

63 | Loglinear models with latent variables. - Hagenaars - 1993 |

49 | Linear logistic latent class analysis for polytomous data - Formann - 1992 |

46 | A general methodology for the analysis of experiments with repeated measurement of categorical data - Koch, Landis, et al. - 1977 |

46 | On the effect of the number of quadrature points in a logistic random effects model: an example - Lesaffre, Spiessens |

45 | Latent class factor and cluster models, bi-plots, and related graphical displays. Sociological Methodology
- Magidson, Vermunt
- 2001
(Show Context)
Citation Context ...06, .67). [INSERT FIGURE 1 ABOUT HERE] Since these factor scores have a distinct probabilistic interpretation, this bi-plot represents an improvement over traditional biplots and perceptual maps (see =-=Magidson and Vermunt 2001-=-). Individual cases can also be plotted based on their factor scores. In addition, LC factor analysis can be performed using fewer variables than traditional factor analysis. In traditional factor ana... |

42 | A General Maximum Likelihood Analysis of Overdispersion in Generalized Linear Models - Aitkin - 1996 |

39 | Latent class models for clustering: a comparison with K-means - Magidson, Vermunt |

38 | A comparison of segment retention criteria for finite mixture logit models. - Andrews, Currim - 2003 |

35 |
A mixed-effects multinomial logistic regression model.
- Hedeker
- 2003
(Show Context)
Citation Context ... application of CFactors involves random-effects modeling using any of the generalized linear models implemented in the Regression 85 Module (Agresti, Booth, Hobert, and Caffo, 2000; Goldstein, 1995; =-=Hedeker, 2003-=-; Hedeker and Gibbons, 1996; Snijders and Bosker, 1999; Skrondal and Rabe-Hesketh, 2004; McFadden and Train, 2000; Wong and Mason, 1985).42 Let us first look at the random intercept model. Denoting th... |

34 | Techniques of event history modeling - Blossfeld, Rohwer - 1995 |

34 |
Multiple imputation of industry and occupation codes in census public-use samples using Bayesian logistic regression.
- Clogg, Rubin, et al.
- 1991
(Show Context)
Citation Context ...let priors for the latent and the conditional response probabilities and gamma priors for the Poisson rates, and the second by using inverse-Wishart priors for the error variance-covariance matrices (=-=Clogg et al., 1991-=-; Galindo-Garre, Vermunt, and Bergsma, 2004: Gelman et. al., 1996; Schafer, 1997). These are so-called conjugate priors since they have the same form as the corresponding multinomial, Poisson, and mul... |

34 |
Concomitant Variable Latent Class Models for the External Analysis of Choice Data," Research Memorandum 486,
- Kamakura, Wedel, et al.
- 1992
(Show Context)
Citation Context ...is restricted by means of a logistic, log-linear, or linear regression model.19 An important extension of the above LC Regression model is obtained by making class membership dependent on covariates (=-=Kamakura, Wedel, and Agrawal, 1994-=-; Vermunt, 1997). An example of such a model is: f(yi|zcovci1 zcovi2 , zpredi1 , zpredi2 ) = K∑ x=1 P (x|zcovi1 , zcovi2 )f(yi|x, zpredi1 , zpredi2 ) In this model, it is assumed that the probability ... |

32 | Fitting multivariate normal finite mixtures subject to structural equation modeling. - Dolan, Maas - 1998 |

32 | A comparison of Poisson, negative binomial, and semiparametric mixed Poisson regression models with applications to criminal careers data. - Land, McCall, et al. - 1996 |

32 | Bootstrapping goodness-of-fit measures in categorical data analysis. - Langeheine, Pannekoek, et al. - 1996 |

31 | Random-effects modeling of categorical response data
- Agresti, Booth, et al.
- 2000
(Show Context)
Citation Context ...ted to zero. 9.2.4 Random-effects models Another important application of CFactors involves random-effects modeling using any of the generalized linear models implemented in the Regression 85 Module (=-=Agresti, Booth, Hobert, and Caffo, 2000-=-; Goldstein, 1995; Hedeker, 2003; Hedeker and Gibbons, 1996; Snijders and Bosker, 1999; Skrondal and Rabe-Hesketh, 2004; McFadden and Train, 2000; Wong and Mason, 1985).42 Let us first look at the ran... |

28 | Latent trait and latent class models. - Langheine, Rost - 1988 |

28 | Latent Class Analysis.” - Vermunt, Magidson - 2003 |

27 |
Latent Structure and Other Mixture Models in Marketing: An Integrative Survey and Overview."
- William, Kumar
- 1992
(Show Context)
Citation Context ...ng one or more discrete unobserved variables. In the context of marketing research, one will typically interpret the categories of these latent variables, the latent classes, as clusters or segments (=-=Dillon and Kumar 1994-=-; Wedel and Kamakura 1998). Among other uses, LC analysis provides a powerful new tool to identify important market segments in target marketing. Recently, the close connection between LC models and r... |

27 |
A review of recent developments in latent class regression models.
- Wedel, DeSarbo
- 1994
(Show Context)
Citation Context ...git model (Andrews et al., 2002; Louviere et al., 2000; McFadden and Train, 2000). The LC choice model can also be seen as a variant of the LC or mixture regression model (Vermunt and Magidson, 2000; =-=Wedel and DeSarbo, 1994-=-, 2002). Most studies will contain multiple observations or multiple replications per respondent: e.g., respondents indicate their first choice for several sets of products or provide ratings for vari... |

27 |
Latent structure models with direct effects between indicators: Local dependence models.
- Hagenaars
- 1988
(Show Context)
Citation Context ...ting strategy that we would like to propagate is to relax the local independence assumption by allowing for associations between indicators, as well as direct effects of covariates on the indicators (=-=Hagenaars, 1988-=-; Vermunt, 1997). Latent GOLD calculates bivariate z–y and y–y residuals which can be used to detect which pairs of observed variables are more strongly related than can be explained by the formulated... |

26 | Goodness-of-fit testing for latent class models - Collins, Fidler, et al. - 1993 |

26 | Latent class scaling analysis. Thousand Oaks, - Dayton - 1998 |

24 |
Mixture model clustering using the MULTIMIX program
- Hunt, Jorgensen
- 1999
(Show Context)
Citation Context ...(6), with Clusterspecific densities of the form described in equation (3). 3.6 LC Cluster Models for Mixed Mode Data The most general LC Cluster model is the model for mixed mode data (Everitt, 1988; =-=Hunt and Jorgensen, 1999-=-; Lawrence and Krzanowski, 1996; Moustaki, 1996; Vermunt and Magidson, 2002). This model is used when one has y variables of different scale types. The structure that serves as the starting point is a... |

22 | Modification indices for the 2PL and the nominal response model. - Glas - 1999 |

21 |
Latent class analysis with ordered latent classes.
- Croon
- 1990
(Show Context)
Citation Context ...stricted Clusters. With this constraint, cluster-specific item probabilities and means will be restricted to be monotonically increasing. It yields what is usually referred to as ordinal LC analysis (=-=Croon, 1990-=-, 2002; Vermunt, 2001). For indicators which are specified to be ordinal, continuous, or counts, the order-restricted-clusters constraint implies the following inequality on the regression parameters:... |

20 |
New developments in latent structure analysis
- Clogg
- 1981
(Show Context)
Citation Context ...adjacent-category ordinal logit model for response variable yit. 3.3 Covariates An important extension of the standard LC model described above is obtained by the possibility of including covariates (=-=Clogg, 1981-=-; Dayton and McReady, 1988; Hagenaars, 1990, 1993).9 Inclusion of covariates to predict class membership is straightforward within the general framework of the model defined in equation (1). Suppose w... |

20 | Generalizing logistic regression by nonparametric mixing - DA, Lambert - 1989 |

19 | A biomedical application of latent class models with random effects,” - Hadgu, Qu - 1998 |

18 |
Latent GOLD 2.0 User’s Guide
- Vermunt, Magidson
- 2000
(Show Context)
Citation Context ...is performed to relate the resulting clusters or factors obtained from a traditional cluster or factor analysis to demographic and other variables. In addition, LC models have recently been extended (=-=Vermunt and Magidson, 2000-=-, 2002) to include variables of mixed scale types (nominal, ordinal, continuous and/or count variables) in the same analysis. Kinds of Latent Class Models Three common statistical application areas of... |

18 |
A finite mixture model for the clustering of mixed-mode data,
- Everitt
- 1988
(Show Context)
Citation Context ...ed in equation (6), with Clusterspecific densities of the form described in equation (3). 3.6 LC Cluster Models for Mixed Mode Data The most general LC Cluster model is the model for mixed mode data (=-=Everitt, 1988-=-; Hunt and Jorgensen, 1999; Lawrence and Krzanowski, 1996; Moustaki, 1996; Vermunt and Magidson, 2002). This model is used when one has y variables of different scale types. The structure that serves ... |

17 | A stabilized Newton-Raphson algorithm for log-linear models for frequency tables derived by indirect observation, - Haberman - 1988 |

12 | A nonparametric random-coefficients approach: the latent class regression model. - unknown authors - 2001 |

12 | Mixed-effects analyses of rank-ordered data. - Böckenholt - 2001 |

11 | Latent class models for measurement. In: - Clogg - 1988 |

11 |
Mixed models for binomial data with an application to lamb mortality
- Im, Gianola
- 1988
(Show Context)
Citation Context ...andom-effects GLMs Combining the GCFactors from the multilevel model with the CFactors option makes it possible to specify “standard” three-level GLM regression models with parametric random effects (=-=Im and Gionala, 1988-=-; Skrondal and Rabe-Hesketh, 2004; Rodriguez and Goldman, 2001; Vermunt, 2002c, 2004). In terms of probability structure, this yields: f(yj|zj) = ∫ Fgj f(Fgj ) Ij∏ i=1 ∫ Fji f(Fji) Ti∏ t=1 f(yjit|z... |

11 |
Maximum likelihood assessment of clinical trials based on an ordered categorical response
- Magidson
- 1996
(Show Context)
Citation Context ...th the average (geometric mean) of the probabilities of all M categories. 14 With an ordinal dependent variable we make use of the adjacent-category ordinal logit model (Agresti, 2002; Goodman, 1979; =-=Magidson, 1996-=-) in which ηm|zi = βm0 + P∑ p=1 β·p · y∗m · zip, irrespective of whether one uses dummy or effect coding for the dependent variable. As can be seen, compared to the nominal logit model, this ordinal l... |

10 | Bangladesh Fertility Survey - Huq, Cleland - 1990 |

9 | Finite Mixture Models: Review, Applications and Computer Intensive Methods,” doctoral dissertation, - Dias - 2004 |

9 | Bayesian posterior estimation of logit parameters with small samples.
- Galindo-Garre, Vermunt, et al.
- 2004
(Show Context)
Citation Context ...atent and the conditional response probabilities and gamma priors for the Poisson rates, and the second by using inverse-Wishart priors for the error variance-covariance matrices (Clogg et al., 1991; =-=Galindo-Garre, Vermunt, and Bergsma, 2004-=-: Gelman et. al., 1996; Schafer, 1997). These are so-called conjugate priors since they have the same form as the corresponding multinomial, Poisson, and multivariate normal probability densities. The... |

8 | 2003. ‘‘Comparing Latent Class Factor Analysis With the Traditional Approach - Magidson, Vermunt |

8 | Ordering the classes - Croon - 2002 |

8 | Analysis of Qualitative Data, Vol 2 - Haberman - 1979 |

8 |
Mixture separation for mixed-mode data
- Lawrence, Krzanowski
- 1996
(Show Context)
Citation Context ...densities of the form described in equation (3). 3.6 LC Cluster Models for Mixed Mode Data The most general LC Cluster model is the model for mixed mode data (Everitt, 1988; Hunt and Jorgensen, 1999; =-=Lawrence and Krzanowski, 1996-=-; Moustaki, 1996; Vermunt and Magidson, 2002). This model is used when one has y variables of different scale types. The structure that serves as the starting point is again the local independence str... |

8 | Discrete-time bivariate hazards with unobserved heterogeneity: A partially observed contingency table approach. - Mare - 1994 |

8 |
Qualitative Variance, Entropy, and Correlation Ratios for Nomina) Dependent Variables
- Magidson
- 1981
(Show Context)
Citation Context ...tative variance labelled R2x,variance. The former is similar to the association measure Lambda and the latter to the Goodman and Kruskal tau-b association coefficient for nominal dependent variables (=-=Magidson, 1981-=-). The proportion of classification errors is defined as: E = ∑I i=1wi [ 1−max P̂ (x|zi,yi) ] N . Each of the three R2x measures is based on the same type of reduction of error structure; namely, R2x ... |

7 | Latent class models for classification. Comput Stat Data Anal 2003;41:531–7 - JK, Magidson |

7 | A latent class regression approach for the analysis of recurrent choice data. - Bockenholt - 1993 |

7 | Bayesian Inference for Finite Mixture Models of Generalized Linear Models with Random Effects,” - Lenk, DeSarbo - 2000 |

7 | Nontechnical Introduction to Latent Class Models - Magidson, Vermun - 2002 |

6 | The goodness-of-fit of latent trait models in attitude measurement. - Bartholomew, Tzamourani - 1999 |

6 | Using Latent Class Models to Analyze Response Patterns - Kohlmann, Formann - 1997 |

6 | Latent class modelling as a probabilistic extension of K-means clustering - Magidson, Vermunt - 2002 |

5 | LADI: A latent discriminant model for analyzing marketing research data - Dillon, Mulani - 1989 |

5 | Structural latent class models - Formann, Kohlmann - 1998 |

5 | Likelihood-ratio tests for order-restricted log-linear models: A comparison of asymptotic and bootstrap methods.Metodologa de las Ciencias del Comportamiento 4(1 - Garre, Vermunt, et al. - 2002 |

5 | MIXPREG: a computer program for mixed-effect Poisson regression - Hedeker - 1998 |

4 | Multilevel Modeling. Thousand Oaks: Sage - Luke - 2004 |

3 |
Statistics and neural networks: advances at the interface
- Kay, Titterington
- 2000
(Show Context)
Citation Context ...ck-box-like” nature, recent advances by statisticians promise similar LC applicationsin the near future that provide more efficient and speedier estimation, and more easily interpretable results (see =-=Kay and Titterington, 1999-=-, Vermunt and Magidson, 2002b). For example, consider the case of nonlinear response models where say response or net revenue from a direct marketing effort is the dependent variable, a “supervised” l... |

3 |
2001. “Latent Class Cluster Analysis”, Chapter 3 in Applied Latent Class Analysis. edited by J.A
- Vermunt, Magidson
(Show Context)
Citation Context ..., 2002; Agresti, 2002). In addition, the close connection between latent classes and “hidden layer nodes” in the most widely used neural net model, the multilayer perceptron (MLP) has been clarified (=-=Vermunt and Magidson, 2002-=-).. These recent developmentsopen the door to the use of latent class models for nonlinear regression applications, providing improvements over the current approaches to both REM and MLP in speed and ... |

3 | Computer-Assisted Analysis of Mixtures: Meta-Analysis, Disease Mapping and Others - Böhning - 2000 |

3 | MIXOR: A computer program for mixed effects ordinal regression analysis - D - 1996 |

2 |
forthcoming). Latent GOLD Choice 3.0 User's Guide
- Vermunt, Magidson
- 2003
(Show Context)
Citation Context ...en, 1986). LC choice models provide an important advance in these models, allowing for different utilities to be estimated for each latent segment (McFadden and Train, 2000, Greene and Hensher, 2002, =-=Vermunt and Magidson, 2003-=-). Like traditional regression modeling, LC regression requires a computer program. As LC regression modeling is relatively new, very few programs currently exist. Our comparisons between LC regressio... |

2 | Testing log-linear models with inequality constraints: A comparison of asymptotic, bootstrap, and posterior predictive p-values - Galindo-Garre, Vermunt - 2005 |

1 | the National Election Studies. NATIONAL ELECTION STUDIES, 2000: PRE-/POST- ELECTION STUDY [dataset id:2000.T]. Ann Arbor - Burns, Kinder, et al. - 2001 |