izpis_h1_title_alt

Simple reparameterization to improve convergence in linear mixed models
Gorjanc, Gregor (Author), Flisar, Tina (Author), Martínez-Ávila, Jose Carlos (Author), García-Cortés, Luis Alberto (Author)

URLURL - Presentation file, Visit http://aas.bf.uni-lj.si/zootehnika/96-2010/PDF/96-2010-2-69-73.pdf This link opens in a new window

Abstract
Slow convergence and mixing are one of the main problems of Markov chain Monte Carlo (McMC) algorithms applied to mixed models in animal breeding. Poor convergence is to a large extent caused by high posterior correlation between variance components and solutions for the levels of associated effects. A simple reparameterization of the conventional model for variance component estimation is presented which improves McMC sampling and provides the same posterior distributions as the conventional model. Reparameterization is based on the rescaling of hierarchical (random) effects in a model, which alleviates posterior correlation. The developed model is compared against the conventional model using several simulated data sets. Results show that presented reparameterization has better behaviour of associated sampling methods and is several times more efficient for the low values of heritability.

Language:English
Keywords:statistics, mixed model, Bayesian analysis, McMC, reparameterization, convergence
Work type:Not categorized (r6)
Tipology:1.01 - Original Scientific Article
Organization:BF - Biotechnical Faculty
Year:2010
Number of pages:Str. 69-73
Numbering:Letn. 96, št. 2
UDC:519.2
ISSN on article:1581-9175
COBISS.SI-ID:2792328 This link opens in a new window
Views:1242
Downloads:509
Metadata:XML RDF-CHPDL DC-XML DC-RDF
 
Average score:(0 votes)
Your score:Voting is allowed only to logged in users.
:
Share:AddThis
AddThis uses cookies that require your consent. Edit consent...

Record is a part of a journal

Title:Acta agriculturae Slovenica
Shortened title:Acta agric. Slov.
Publisher:Biotehniška fakulteta
ISSN:1581-9175
COBISS.SI-ID:213840640 This link opens in a new window

Secondary language

Language:Slovenian
Abstract:
Počasna konvergenca je eden največjih problemov uporabe metode Monte Carlo z Markovimi verigami (McMC) za mešane modele na področju genetike in selekcije domačih živali. Slaba konvergenca je v veliki meri posledica visoke posteriorne korelacije med komponentami variance in rešitvami za ravni pripadajočih vplivov. Predstavljamo enostavno reparametrizacijo običajnega modela, ki izboljša lastnosti metode McMC in daje enake posteriorne porazdelitve parametrov modela kot standardni pristop. Reparametrizacija temelji na standardizaciji hierarhičnih (naključnih) vplivov v modelu, kar posledično spremeni posteriorne korelacije med parametri. Oba pristopa smo primerjali na večjem setu simuliranih podatkov. Rezultati kažejo, da reparametrizacija vodi do bolj učinkovitih metod McMC vzorčenja in je nekajkrat bolj učinkovita za analizo lastnosti z nizko heritabilitet.

Keywords:statistika, mešani model, bayesovska analiza, McMC, reparametrizacija, konvergenca

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Comments

Leave comment

You have to log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back