scholarly journals Statistical Analysis of Mixed Model Factorial Experiments with Missing Factor Combinations: The Case of Asynchronous Cyclic Drought Data

2006 ◽  
Vol 131 (2) ◽  
pp. 201-208
Author(s):  
Dawn M. VanLeeuwen ◽  
Rolston St. Hilaire ◽  
Emad Y. Bsoul
Keyword(s):  
Statistical Analysis ◽  
Least Squares ◽  
Mixed Models ◽  
Repeated Measures ◽  
Fixed Effects ◽  
Information Criteria ◽  
Covariance Structures ◽  
Data Set ◽  
Nested Models ◽  
The Cross

Statistical analysis of data from repeated measures experiments with missing factor combinations encounters multiple complications. Data from asynchronous cyclic drought experiments incorporate unequal numbers of drought cycles for different sources and provide an example of data both with repeated measures and missing factor combinations. Repeated measures data are problematic because typical analyses with PROC GLM do not allow the researcher to compare candidate covariance structures. In contrast, PROC MIXED allows comparison of covariance structures and several options for modeling serial correlation and variance heterogeneity. When there are missing factor combinations, the cross-classified model traditionally used for synchronized trials is inappropriate. For asynchronous data, some least squares means estimates for treatment and source main effects, and treatment by source interaction effects are inestimable. The objectives of this paper were to use an asynchronous drought cycle data set to 1) model an appropriate covariance structure using mixed models, and 2) compare the cross-classified fixed effects model to drought cycle nested within source models. We used a data set of midday water potential measurements taken during a cyclic drought study of 15 half-siblings of bigtooth maples (Acer grandidentatum Nutt.) indigenous to Arizona, New Mexico, Texas, and Utah. Data were analyzed using SAS PROC MIXED software. Information criteria lead to the selection of a model incorporating separate compound symmetric covariance structures for the two irrigation treatment groups. When using nested models in the fixed portion of the model, there are no missing factors because drought cycle is not treated as a crossed experimental factor. Nested models provided meaningful F tests and estimated all the least squares means, but the cross-classified model did not. Furthermore, the nested models adequately compared the treatment effect of sources subjected to asynchronous drought events. We conclude that researchers wishing to analyze data from asynchronous drought trials must consider using mixed models with nested fixed effects.

scholarly journals Maybe maximal: Good enough mixed models optimize power while controlling Type I error

2019 ◽  
Author(s):  
Michael Seedorff ◽  
Jacob Oleson ◽  
Bob McMurray
Keyword(s):  
Random Effects ◽  
Mixed Models ◽  
Repeated Measures ◽  
Fixed Effects ◽  
Degrees Of Freedom ◽  
Linear Models ◽  
Type I Error ◽  
Logistic Models ◽  
Information Criteria ◽  
Type I

Mixed effects models have become a critical tool in all areas of psychology and allied fields. This is due to their ability to account for multiple random factors, and their ability to handle proportional data in repeated measures designs. While substantial research has addressed how to structure fixed effects in such models there is less understanding of appropriate random effects structures. Recent work with linear models suggests the choice of random effects structures affects Type I error in such models (Barr, Levy, Scheepers, & Tily, 2013; Matuschek, Kliegl, Vasishth, Baayen, & Bates, 2017). This has not been examined for between subject effects, which are crucial for many areas of psychology, nor has this been examined in logistic models. Moreover, mixed models expose a number of researcher degrees of freedom: the decision to aggregate data or not, the manner in which degrees of freedom are computed, and what to do when models do not converge. However, the implications of these choices for power and Type I error are not well known. To address these issues, we conducted a series of Monte Carlo simulations which examined linear and logistic models in a mixed design with crossed random effects. These suggest that a consideration of the entire space of possible models using simple information criteria such as AIC leads to optimal power while holding Type I error constant. They also suggest data aggregation and the d.f, computation have minimal effects on Type I Error and Power, and they suggest appropriate approaches for dealing with non-convergence.

scholarly journals Anpassung eines Fixed Regression Modells für die tägliche Zunahme von Fleckviehbullen mit Hilfe von Informationskriterien

2007 ◽  
Vol 50 (1) ◽  
pp. 47-58
Author(s):  
N. Mielenz ◽  
H. Krejčová ◽  
J. Přibyl ◽  
L. Schüler
Keyword(s):  
Fixed Effects ◽  
Covariance Structure ◽  
Information Criterion ◽  
Complex Model ◽  
Second Order ◽  
Information Criteria ◽  
Covariance Structures ◽  
Expected Gain ◽  
Bayesian Criteria ◽  
Daily Gain

Abstract. Title of the paper: Fitting a fixed regression model for daily gain of bulls using information criterion In this study the model choice is demonstrated exemplarily on data of 6405 Czech Simmental bulls using information criterion. Per bull up to 8 observations were available for the trait daily gain. Because the animals showed different age on control day, the expected gain curves were described in the population and within the herd*year*season-classes by second, third or fourth order Legendre polynomials of age. For optimization of the fixed effects and to choice the covariance structure of the repeated records the information criteria of Akaike (AIC), the Bayesian criteria (BIC) and the ICOMP-criteria, developed mainly from Bozdogan, were used. Within and over all covariance structures AIC selected generally the most complex model. On the other hand, BIC and ICOMP favoured a model with second order polynomials of age nested within the head*year*seasonclasses. All criterion selected models with nested second order polynomials within the herd*year*season-classes in comparison to models with non-nested polynomials of age.

scholarly journals Analysis of genetic parameters and genetic trends for early growth traits in Iranian Afshari sheep

2018 ◽  
Vol 34 (3) ◽  
pp. 289-301
Author(s):  
Meysam Latifi ◽  
Ali Mohammadi
Keyword(s):  
Genetic Improvement ◽  
Fixed Effects ◽  
Genetic Parameters ◽  
Growth Traits ◽  
Early Growth ◽  
Information Criteria ◽  
Daily Weight Gain ◽  
Data Set ◽  
Female Birth ◽  
Genetic Trends

The purpose of the present study was estimation of genetic parameters and genetic trends of early growth traits using Bayesian approach by Gibbs3f90 software in Iranian Afshari sheep. The data set [birth weight (BW), weaning weight (WW) and pre-weaning daily weight gain (PWDG)] were collected during the period 1999 to 2010 from Agriculture Jahad of Zanjan province, Iran. The fitted fixed effects were herd-year-season as interactions, sex (male, female), birth type (single, multiple) and age of dam. Based on Derivative Information Criteria (DIC), for studied traits the most appropriate model was determined. Therefore, based on the most appropriate fitted model, the direct additive heritabilities estimate for BW, WW and PWDG were 0.32?0.02, 0.05?0.01 and 0.24?0.02, respectively. The estimates of maternal heritabilities were 0.17?0.04, 0.07?0.02 and 0.12?0.05 and total heritabilities 0.11?0.05, 0.08?0.02 and 0.08?0.03 for BW, WW and PWDG, respectively. Direct genetic trends were positive for all traits but only significant for BW 0.75?0.31 g/year (P <0.05). Also, maternal genetic trends were for all traits negative and was significant for BW -0.63?0.27 g/year (P <0.05). The moderate estimates of heritabilities for early growth traits indicate that in Afshari sheep faster genetic improvement through selection is possible for these traits. Furthermore, the results genetic trends in this current study indicated that genetic improvement through selection is suitable only for BW in Afshari sheep.

scholarly journals EFEITO DO MEIO E DA IDADE DA VACA SOBRE O GANHO DE PESO NA FASE DE CRIA, RECRIA E OS DIAS PARA SE OBTER 160 KG EM ANIMAIS DA RAÇA GUZERÁ

2004 ◽  
Vol 9 (1) ◽  
Author(s):  
E. GEMIN ◽  
J.C. SOUZA ◽  
L.O.C. SILVA ◽  
C.H.M. MALHADO ◽  
P.B. FERRAZ FILHO
Keyword(s):  
Environmental Factors ◽  
Statistical Model ◽  
Least Squares ◽  
Random Effects ◽  
Environmental Effects ◽  
Fixed Effects ◽  
Least Squares Method ◽  
Data Set ◽  
Quadratic Effect ◽  
Daily Gain

O objetivo deste trabalho foi avaliar a influência dos efeitos de meio e da idade da vaca sobre os ganhos de peso do nascimento ao desmame (GPD), período pós-desmame (GPS) e sobre o número de dias para se obter 160 kg (D160). O rebanho avaliado continha 1.747 animais, sendo os dados analisados pelo método dos quadrados mínimos utilizando-se um modelo estatístico contendo os efeitos fixos de mês, ano e sexo do bezerro, o efeito aleatório de touros na fazenda, e como covariável a idade da vaca ao parto. As médias ajustadas para ganho de peso pré e pós-desmame, e para dias para a obtenção 160 kg foram 0,604 ± 0,01 kg; 0,399 ± 0,01 kg; em 285 ± 5,3 dias, respectivamente. Os machos foram superiores às fêmeas relativo ao GPD = 6,0%; D160 = 5,8 %, GPS = 20,1%. Quanto ao mês, as maiores médias de ganho de peso no pré-desmame recaiu nos animais nascidos no mês de agosto. Com relação aos dias para se obter 160 kg, os melhores resultados foram dos animais nascidos nos meses julho a setembro. A idade da vaca influenciou as caracteristicas ganho de peso pré-desmame e no D160. Environmental effects and age of dam on pre- and post-weaning daily gain, and on number of days to gain 160 kg from birth to weaning on guzerath breed cattle Abstract The objective of this study was to evaluate the effects of environmental factors and age of dam on pre- (GPD) and post-weaning (GPS) daily gain, and on the were number of days to gain 160 kg (D160) from birth to weaning. The data set contained 1,747 animals, and were analyzed by the least squares method. The statistical model included the fixed effects of month and year of birth, and sex of the calf and age of the dam at calving. Sire nested within farm and the error were random effects. The pre- and post-weaning average daily gains, and days to gain 160 kg least squares means were 0.604 ± 0.01 kg, 0.399 ± 0.01 kg, and 285.0 ± 5.3 days, respectively. The males were 6.0, 21.1 and 5.8% superior to the females for GPD, GPS and D160, respectively. The highest pre-weaning gain was for the animals born August. Regarding D160, the best results were for the animals born from July to September. Age of the cow showed a significant quadratic effect on the traits. The best cows were the 94-month-old ones. First calving cows produced the lightest calves. The results showed the importance of the environmental effects on the traits studied, evidencing the need for them to be corrected.

scholarly journals R2s for Correlated Data: Phylogenetic Models, LMMs, and GLMMs

2017 ◽  
Author(s):  
Anthony R. Ives
Keyword(s):  
Logistic Regression ◽  
Least Squares ◽  
Mixed Models ◽  
Fixed Effects ◽  
Ordinary Least Squares ◽  
Correlated Data ◽  
Coefficient Of Determination ◽  
Advantages And Disadvantages ◽  
Phylogenetic Models ◽  
Variance Explained

AbstractMany researchers want to report an R2 to measure the variance explained by a model. When the model includes correlation among data, such as phylogenetic models and mixed models, defining an R2 faces two conceptual problems. (i) It is unclear how to measure the variance explained by predictor (independent) variables when the model contains covariances. (ii) Researchers may want the R2 to include the variance explained by the covariances by asking questions such as “How much of the variance is explained by phylogeny?” Here, I investigate three R2s for phylogenetic and mixed models. A least-squares R2ls is an extension of the ordinary least-squares R2 that weights residuals by variances and covariances estimated by the model; it is closely related to R2glmm proposed by Nakagawa & Schielzeth (2013). The conditional expectation R2ce is based on “predicting” each residual from the remaining residuals of the fitted model. The likelihood ratio R2lr was first used by Cragg & Uhler (1970) for logistic regression, and here is used with the standardization proposed by Nagelkerke (1991). These three R2s are formulated as partial R2s, making it possible to compare the contributions of mean components (regression coefficients in phylogenetic models and fixed effects in mixed models) and variance components (phylogenetic correlations and random effects) to the fit of models. The properties of the R2s for phylogenetic models were assessed using simulations for continuous and binary response data (phylogenetic generalized least squares and phylogenetic logistic regression). Because the R2s are designed broadly for any model for correlated data, the R2s were also compared for LMMs and GLMMs. R2ls, R2ce, and R2lr all have good performance, and each has advantages and disadvantages for different applications. These R2s are computed in the R package rr2 (https://github.com/arives/rr2). [Binomial regression, coefficient of determination, non-independent residuals, phylogenetic model, pseudo-likelihood]

scholarly journals Modeling Count Data from Dependent Clusters with Poisson Mixed Models

2018 ◽  
pp. 1-21
Author(s):  
K. A. N. K. Karunarathna ◽  
Pushpakanthie Wijekoon
Keyword(s):  
Mixed Models ◽  
Fixed Effects ◽  
Real Data ◽  
Standard Errors ◽  
Estimation Of Parameters ◽  
Data Set ◽  
New Approach ◽  
Simulation Process ◽  
Cluster Set ◽  
Set Up

Responses collected from dependent clusters are affected by the dependence among clusters and it should be taken into account in modeling such responses. In this study, a new approach was evaluated to incorporate cluster dependence in generalized linear Poisson mixed models for count responses from dependent clusters. Performance of this approach was evaluated by using a simulation process under three different designs and different covariates. The Marginal Generalized Quasi-likelihood (GQL) method was used for estimation of parameters with the cluster dependence. Monte Carlo likelihood (MCL) and Penalized Quasi-likelihood (PQL) estimates also were obtained for the purpose of comparison.  Proposed approach was tested with a real data set also. The proposed approach, with the incorporation of cluster dependence, gives better estimates for both fixed effects and variance of random effects with low standard errors with compared to the estimates obtained by ignoring the cluster dependence. Therefore, the proposed approach can be used for modeling count responses from a dependent cluster set up.

Maintenance of a Genetic Data Set on Multiple Data Base and Statistical Analysis Systems

1982 ◽  
Vol 61 (s109) ◽  
pp. 34-34
Author(s):  
Samuel J. Agronow ◽  
Federico C. Mariona ◽  
Frederick C. Koppitch ◽  
Kazutoshi Mayeda
Keyword(s):  
Statistical Analysis ◽  
Data Base ◽  
Genetic Data ◽  
Data Set ◽  
Multiple Data

The antecedents of corporate social and environmental responsibility discourse in Pakistan: Multiple theoretical perspectives

2019 ◽  
Author(s):  
Muhammad Farhan Basheer ◽  
Saqib Muneer ◽  
Muhammad Atif ◽  
Zubair Ahmad
Keyword(s):  
Institutional Theory ◽  
Least Squares ◽  
Agency Theory ◽  
Stakeholder Theory ◽  
Fixed Effects ◽  
Ordinary Least Squares ◽  
Environmental Responsibility ◽  
Legitimacy Theory ◽  
Discourse Practices ◽  
Corporate Social

The primary purpose of the study is to explore the antecedents of corporate social and environmental responsibilities discourse practices in Pakistan. The industry sensitivity, government shareholding, block holder ownership, print media coverage, environmental monitoring programs, and strategic posture are examined as antecedents of corporate social and environmental responsibility practices. A multidimensional theoretical perspective namely stakeholder theory (ST), institutional theory (IT), agency theory (PAT), and legitimacy theory (LT) is used to conceptualize the phenomena. All the four of perspective theories (positive accounting theory, legitimacy theory, stakeholder theory, and institutional theory) claim that there are ‘pressures’ that impact the organization. How much ‘pressures’ are recognized, managed or satisfied differs from one perspective of theory to the other. To estimate the data, this study uses three sets of panel data models, i.e., the pooled ordinary least squares model (POLS) or constant coefficients model, fixed effects (FEM or least squares dummy variable/LSDV model) and random-effects models. The final sample is comprising of 173 firms over eight years from 2011 to 2017. The firms listed in PSX are included in the sample. Overall the findings of the study have shown agreement with the proposed results. However, the study has provided more support to the institutional theory and stakeholder theory. Keywords: Corporate Social Responsibility, Stakeholders Theory, Agency Theory, Pakistan

Sign in / Sign up

Export Citation Format

Share Document