Multinomial logit random effects models

Hartzel, J.; Agresti, Alan; Caffo, B.

doi:10.1191/147108201128104

Cited by 105 publications

(104 citation statements)

References 39 publications

Supporting

Mentioning

100

Contrasting

Unclassified

Order By: Relevance

“…We then entered variables into the mixed-effect multinomial logistic model by forward selection with alpha set at 0.10 for variable inclusion or removal (Hartzel et al 2001;Sergio et al 2003). Social behavior was the dependant variable and had three categories.…”

Section: Statistical Analysesmentioning

confidence: 99%

Winter space use and social behaviors of Chinese Grouse (Bonasa sewerzowi) at Lianhuashan mountains, Gansu, China

2010

View full text Add to dashboard Cite

Winter movement and social behavior of Chinese Grouse (Bonasa sewerzowi) were studied at Lianhuashan Natural Reserve, northwest China from 1995 to 2009. According to radio-tracking data from 25 male Chinese Grouse, their home ranges in winter ranged from 6.0 ± 1.6 ha (n = 9) to 16.0 ± 2.6 ha (n = 10). Most females and all the yearlings moved from breeding areas and formed mixed-sex flocks. However, males adopted one of three social behaviors and movement patterns, including: staying in the territory, moving to the winter flocking area, or wandering between the spring territory and the wintering area. In order to find out which factors influence male social behavior, we used multinomial logistic regression and AIC model selection to compare the habitat characteristics of male Chinese Grouse breeding territories. Modeling results indicated that food, cover, topography, and future breeding opportunity were the main variables determining social behavior in winter. KeywordsBonasa sewerzowi Á Social behaviors Á Non-breeding season Á Trade-off Á Qinghai-Tibetan plateau Zusammenfassung Winterbewegungen und Sozialverhalten von Schwarzbrust-Haselhühnern (Bonasa sewerzowi) wurden im Lianhuashan-Naturschutzgebiet in Nordwestchina von 1995 bis 2008 untersucht. Den Radiotelemetriedaten von 25 männlichen Schwarzbrust-Haselhühnern zufolge reichten ihre Streifgebiete im Winter von 6,04 ± 1,60 ha (n = 9) bis 15,98 ± 2,61 ha (n = 10). Die meisten Weibchen sowie alle Jährlinge entfernten sich von den Brutgebieten und bildeten gemischtgeschlechtliche Schwärme. Bei Männchen ließen sich jedoch drei verschiedene soziale Verhaltensweisen und Bewegungsmuster beobachten, nämlich Verbleiben im Territorium, Wegzug ins Winterschwarmgebiet und Wechseln zwischen dem Frühjahrsterritorium und dem Ü berwinterungsgebiet. Um herauszufinden, welche Faktoren das Sozialverhalten von Männchen beeinflussen, haben wir zum Vergleich der Habitateigenschaften der Brutreviere männlicher Schwarzbrust-Haselhühner multinomiale logistische Regression und AIC-Modellselektion angewendet. Die Modellierungsergebnisse deuteten darauf hin, dass Nahrung, Deckung, Topographie und zukünftige Brutgelegenheiten die Hauptvariablen waren, die das Sozialverhalten im Winter bestimmten.

show abstract

Section: Statistical Analysesmentioning

confidence: 99%

Winter space use and social behaviors of Chinese Grouse (Bonasa sewerzowi) at Lianhuashan mountains, Gansu, China

2010

View full text Add to dashboard Cite

show abstract

“…Multinomial random effects models have previously been treated as special cases of multivariate generalized linear mixed models (MGLMMs) (Tutz and Hennevogl 1996;Hartzel et al 2001). We also adopt this general approach as it provides unified fitting and inferential procedures for a broad class of models.…”

mentioning

confidence: 99%

“…Nonetheless, the NP approach yields fairly efficient estimates of the effects of the explanatory variables and it has been used extensively by statisticians. For instance, Aitkin (1999) used this approach in the context of generalized linear models and Hartzel et al (2001) for modeling correlated multinomial responses. General fitting algorithms have been provided by Laird (1978), Lindsay (1983), Follmann andLambert (1989) and Lesperance and Kalbfleisch (1992).…”

mentioning

confidence: 99%

Multivariate generalized linear mixed models with semi-nonparametric and smooth nonparametric random effects densities

Παπαγεωργίου

Hinde

2010

Stat Comput

View full text Add to dashboard Cite

We extend the family of multivariate generalized linear mixed models to include random effects that are generated by smooth densities. We consider two such families of densities, the so-called semi-nonparametric (SNP) and smooth nonparametric (SMNP) densities. Maximum likelihood estimation, under either the SNP or the SMNP densities, is carried out using a Monte Carlo EM algorithm. This algorithm uses rejection sampling and automatically increases the MC sample size as it approaches convergence. In a simulation study we investigate the performance of these two densities in capturing the true underlying shape of the random effects distribution. We also examine the implications of misspecification of the random effects distribution on the estimation of the fixed effects and their standard errors. The impact of the assumed random effects density on the estimation of the random effects themselves is investigated in a simulation study and also in an application to a real data set.

show abstract

“…Several further additions and variants may be included in these models and we quote the possibility to express the effects of subjects' covariates on the shelter choice (as pursued by Iannario and Piccolo 2010b in a real case study) or to extend our proposal in a multilevel framework (Iannario 2011b), as it would be convenient for assessing variations among and within subgroups of ordinal data sets (see Hartzel et al 2001;Grilli and Rampichini 2002;Fielding et al 2003, among others). number E61J10000020001, both within the Research Unit at University of Naples Federico II.…”

Section: Discussionmentioning

confidence: 99%

Modelling shelter choices in a class of mixture models for ordinal responses

Iannario

2011

Stat Methods Appl

View full text Add to dashboard Cite

In rating surveys, people are requested to evaluate objects, items, services, and so on, by choosing among a list of ordered categories. In some circumstances, it may happen that a subset of respondents selects a specific option just to simplify a more demanding choice. In this context, we generalize a class of ordinal data models (called CUB and proven effective for fitting and interpretation), for taking the possible presence of a shelter choice into account. After the discussion of interpretative and inferential issues, the usefulness of the approach is checked against real case studies and by means of a simulation experiment. Some final remarks end the paper

show abstract

Multinomial logit random effects models

Cited by 105 publications

References 39 publications

Winter space use and social behaviors of Chinese Grouse (Bonasa sewerzowi) at Lianhuashan mountains, Gansu, China

Winter space use and social behaviors of Chinese Grouse (Bonasa sewerzowi) at Lianhuashan mountains, Gansu, China

Multivariate generalized linear mixed models with semi-nonparametric and smooth nonparametric random effects densities

Modelling shelter choices in a class of mixture models for ordinal responses

Contact Info

Product

Resources

About