scholarly journals Professor Endrenyi’s Legacy: An Evaluation of the Regulatory Requirement “Fixed Effects, Rather Than Random Effects, Should Be Used for All Terms”

2021 ◽  
Vol 24 ◽  
pp. 413-420
Author(s):  
Anders Fuglsang
Keyword(s):  
Variance Components ◽  
Fixed Effects ◽  
Mixed Model ◽  
Linear Models ◽  
Regulatory Requirement ◽  
Subject Variability ◽  
Acceptance Range ◽  
Biased Estimates ◽  
True Values

Purpose: In the latest revision of the guideline for evaluation of bioequivalence (BE), European regulators introduced the requirement for using subjects as fixed factors in the underlying statistical models, even in replicate and semi-replicate studies. The implication was that estimates of within-subject variability were derived with a linear model rather than with a mixed model based on restricted maximum likelihood (REML). While REML-based methods are generally thought to give rise to less biased estimates of variance components, there have been no studies that compared the quality of REML-based estimates and estimates derived via linear models. Methods: A publication by Endrenyi and Tothfalusi from 1999 described simulations in a fashion that is useful for testing the European Medicines Agency’s (EMA) requirement.  This study defines 7 scenarios within which 10,000 individual 2-sequence, 2-treatment, 4-period trials are simulated and makes a comparison of the quality of estimates. Results: It is concluded that estimates based on REML are closer to the true values than estimates based on linear models, but significant differences are only shown in two of the seven scenarios tested.  REML-based estimators have less variability. Both types of estimates appear negatively biased and will therefore decrease the width of the acceptance range.  

scholarly journals Effects of Breed and Stage of Lactation on Milk Fatty Acid Composition of Italian Goat Breeds

Animals ◽  
2019 ◽  
Vol 9 (10) ◽  
pp. 764 ◽  
Author(s):  
Sarah Currò ◽  
Carmen Manuelian ◽  
Massimo De Marchi ◽  
Salvatore Claps ◽  
Domenico Rufrano ◽  
...  
Keyword(s):  
Fatty Acid ◽  
Repeated Measures ◽  
Dairy Products ◽  
Fixed Effects ◽  
Sensory Quality ◽  
Mixed Model ◽  
Goat Milk ◽  
Stage Of Lactation

Fatty acid (FA) profile plays an important role on human health and on sensory quality of dairy products. There is few information about breed influence on milk FA profile of local goat breeds. This study aimed to characterize and compare the milk FA profile of 5 local endangered goat breeds (Garganica, Girgentana, Jonica, Maltese, and Mediterranean Red) and a cosmopolitan breed (Saanen) reared in the same farm during a complete lactation. A total of 252 milk samples were collected monthly from 42 goats (7 goats per breed) and analyzed for gross composition and FA profile. Individual FA was determined using gas-chromatography. Data were analyzed using a mixed model with repeated measures with breed and week of lactation as fixed effects. Results showed that the FA profile was significantly affected by week of lactation and only few FA by breed effect (p < 0.05). Overall, the main differences were found between Saanen and local breeds. This study contributed to the characterization of goat milk FA profile, and it may be of interest for the valorization of milk from local goat breeds which seem to have a healthier profile than milk of the cosmopolitan breed.

The use of optimal evidence-based chemotherapy (chemo) regimens in physician offices versus hospital outpatient facilities.

2019 ◽  
Vol 37 (15_suppl) ◽  
pp. 6514-6514
Author(s):  
Michael Jordan Fisch ◽  
Ying Liu ◽  
ANN NGUYEN ◽  
John Barron ◽  
Ezra Fishman ◽  
...  
Keyword(s):  
Quality Of Care ◽  
Fixed Effects ◽  
Linear Models ◽  
Hospital Setting ◽  
Administrative Claims ◽  
Cancer Type ◽  
Strong Basis ◽  
Hospital Outpatient ◽  
Colorectal Cancer Patients

6514 Background: The proportion of infused chemo administered in hospital outpatient facilities (HOF) increased from 6% in 2004 to 43% in 2014. The average annual cost for patients receiving chemo was significantly higher in HOFs than in physician offices (POs). One option to explore differences in the quality of care between these two settings is to examine the use of chemo regimens, which, based on their efficacy, toxicity, and costs, have been designated as “on-pathway.” This study compared on-pathway rates among patients receiving infused chemo administered in POs vs. those in HOFs. Methods: Using administrative claims data, we identified 61,496 breast, lung, or colorectal cancer patients receiving chemo from 2013 to 2018. Chemo regimens were considered “on-pathway” when they were on payer's program list of optimal regimens when administered. Generalized linear models examined the association between site of service and on-pathway prescribing rates, and costs of care. Models adjusted for age, sex, year, rural status, cancer type and setting, and comorbidities, with fixed effects for providers. Results: Percentage of infused chemo administered in HOFs increased from 44.2% in 2013 to 54.7% in 2018. After adjustment, on-pathway prescribing rate did not differ significantly between HOFs and POs (50.1%, 95% CI: 48.6%-51.5% vs. 49.8%, 95%CI: 48.3%-51.3%, p = 0.65). 6-month chemo cost ($56,885, 95% CI: $54,364-$59,524 vs $32,240, 95% CI: $30,929-$33,605, p < 0.001) and 6-month medical cost ($114,280, 95% CI: $110,716-$117,960 vs $79,455, 95% CI: $77,089-$81,893, p < 0.001) were significantly higher in HOFs vs. POs. Conclusions: Quality of care as measured by use of optimal chemo regimens was similar in hospital and office setting. Cost continues to be significantly higher in hospital setting. These findings provide a strong basis for site-neutral reimbursement policies.

scholarly journals Parámetros genéticos de crecimiento y producción de lana en ovinos usando la metodología de modelos mixtos

2009 ◽  
Vol 7 (1) ◽  
pp. 42
Author(s):  
Rodrigo Alfredo Martínez ◽  
Rodrigo Vásquez ◽  
Javier Vanegas ◽  
Marco Suárez
Keyword(s):  
Variance Components ◽  
Fixed Effects ◽  
Genetic Parameters ◽  
Mixed Model ◽  
Growth Characteristics ◽  
Random Component ◽  
Wool Production ◽  
Gibbs Sampling Algorithm ◽  
Calving Season ◽  
Number Of Births

<p>En este trabajo se estimaron los parámetros genéticos a partir de datos de campo registrados durante 26 años (1974 a 1999) en un núcleo de las razas ovinas Hampshire, Romney Marsh y Corriedale. Se estimaron los parámetros genéticos de las características de crecimiento, producción de lana y reproducción. Los modelos utilizados incluyeron efectos fijos, como ‘año de parto, ‘época de parto, ‘sexo, ‘número de parto; en el componente aleatorio también se incluyeron los efectos genéticos directo, materno y de ambiente permanente, dependiendo de la característica. Los análisis fueron hechos usando un modelo animal de característica simple mediante el programa MTGSAM® que emplea el algoritmo de muestreo de Gibbs para inferir los componentes de varianza. Se encontraron valores de heredabilidad directa que variaron entre 0,13 y 0,33 para peso al nacimiento y entre 0,16 y 0,35 para peso al destete corregido a 120 días y para caracteres de producción de lana, valores cercanos a 0,25. También se determinaron las tendencias anuales promedio de los valores genéticos y de los índices de consanguinidad para cada una de las razas, encontrando en términos generales valores promedio de consanguinidad de 3,47% para la raza Hampshire, 0,75% para Romney Marsh y 0,76% para Corriedale. Estos resultados sugieren una tendencia levemente creciente de los valores de consanguinidad, pero de acuerdo con lo que se espera para una población de conservación.  </p><p> </p><p><strong>Growth and wool production genetic parameters in sheep using mixed model methodology </strong></p><p>This work was aimed at estimating genetic parameters from field-data recorded over a 26-year period (1974 to 1999) in a flock of Hampshire, Romney Marsh and Corriedale sheep. Genetic parameters were estimated from growth characteristics, wool production and reproduction. The models used included fixed effects, such as birth year, calving season, gender and number of births. Direct, maternal and permanent environmental genetic effects were also included in the random component, depending on the characteristic. MTGSAM® software (using Gibbs’ sampling algorithm) was used on a simple animal model for inferring variance components. Direct heritability values were found, ranging form 0.13 to 0.33 for birth weight, 0.16 to 0.35 for weaning weight (corrected to 120 days) and values around 0.25 for wool production. Annual mean tendency of genetic values and consanguinity indices were also determined for each of the breeds, resulting in 3.47% mean consanguinity for the Hampshire, 0.75% for Romney Marsh and 0.76% for Corriedale breeds. These results suggested a slight tendency for increased consanguinity values, but remaining in line with those expected for a conservation population. </p>

scholarly journals Selection in several environments by BLP as an alternative to pooled anova in crop breeding

2009 ◽  
Vol 33 (5) ◽  
pp. 1342-1350 ◽  
Author(s):  
Júlio Sílvio de Sousa Bueno Filho ◽  
Roland Vencovsky
Keyword(s):  
Plant Breeding ◽  
Random Effects ◽  
Fixed Effects ◽  
Mixed Model ◽  
Linear Models ◽  
Minimum Variance ◽  
Pooled Analysis ◽  
Balanced Designs ◽  
Best Linear Unbiased ◽  
Combine Information

Plant breeders often carry out genetic trials in balanced designs. That is not always the case with animal genetic trials. In plant breeding is usual to select progenies tested in several environments by pooled analysis of variance (ANOVA). This procedure is based on the global averages for each family, although genetic values of progenies are better viewed as random effects. Thus, the appropriate form of analysis is more likely to follow the mixed models approach to progeny tests, which became a common practice in animal breeding. Best Linear Unbiased Prediction (BLUP) is not a "method" but a feature of mixed model estimators (predictors) of random effects and may be derived in so many ways that it has the potential of unifying the statistical theory of linear models (Robinson, 1991). When estimates of fixed effects are present is possible to combine information from several different tests by simplifying BLUP, in these situations BLP also has unbiased properties and this lead to BLUP from straightforward heuristics. In this paper some advantages of BLP applied to plant breeding are discussed. Our focus is on how to deal with estimates of progeny means and variances from many environments to work out predictions that have "best" properties (minimum variance linear combinations of progenies' averages). A practical rule for relative weighting is worked out.

HERITABILITY OF NONRETURN RATE OF CANADIAN HOLSTEIN-FRIESIAN BULLS

1983 ◽  
Vol 63 (1) ◽  
pp. 39-48 ◽  
Author(s):  
B. B. MURRAY ◽  
L. R. SCHAEFFER ◽  
E. B. BURNSIDE
Keyword(s):  
Maximum Likelihood ◽  
Artificial Insemination ◽  
Variance Components ◽  
Fixed Effects ◽  
Mixed Model ◽  
Factors Affecting ◽  
Holstein Friesian ◽  
Best Linear Unbiased ◽  
Dairy Bulls ◽  
One Year

Conception rates of Canadian Holstein-Friesian bulls in artificial insemination were recorded as percent 60- to 90-day nonreturns to first service. A total of 5, 923, 974 first services to 784 bulls by 207 sires were analyzed using a mixed model which included age of bull (1–14 yr), years (1956–1976), months and AI units as fixed effects. Bulls and sires of bulls were random. Significant differences were found in all classifications involving fixed effects. Nonreturn rates at one year of age were highest and significantly different from nonreturn rates at 9 yr of age and older. Lowest estimates of nonreturn rates were in January and December, and highest were in September and October. Differences were found among some AI units. Best linear unbiased predictions of differences among all service sires were obtained. Variance components were estimated iteratively by maximum likelihood. Heritability was 0.25 by regression of sire on son with 30 sires having four or more sons. Repeatability was 0.41. Many factors affecting nonreturn rates could not be identified, especially at the cow or herd level. Sire selection programs in AI units for high nonreturn rates in sires of young bulls should be successful. Key words: Fertility, dairy bulls, heritability

scholarly journals AFEchidna is a R package for genetic evaluation of plant and animal breeding datasets

2021 ◽  
Author(s):  
Weihua Zhang ◽  
Ruiyan Wei ◽  
Yan Liu ◽  
Yuanzhen Lin
Keyword(s):  
Random Effects ◽  
Variance Components ◽  
Fixed Effects ◽  
Genetic Parameters ◽  
Linear Models ◽  
R Package ◽  
Animal Breeding ◽  
Mixed Linear Models ◽  
Best Linear Unbiased ◽  
Genetic Assessment

Progeny tests play important roles in plant and animal breeding programs, and mixed linear models are usually performed to estimate variance components of random effects, estimate the fixed effects (Best Linear Unbiased Estimates, BLUEs) and predict the random effects (Best Linear Unbiased Predictions, BLUPs) via restricted maximum likehood (REML) methods in progeny test datasets. The current pioneer software for genetic assessment is ASReml, but it is commercial and expensive. Although there is free software such as Echidna or the R package sommer, the Echidna syntax is complex and the R package functionality is limited. Therefore, this study aims to develop a R package named AFEchidna based on Echidna software. The mixed linear models are conveniently implemented for users through the AFEchidna package to solve variance components, genetic parameters and the BLUP values of random effects, and the batch analysis of multiple traits, multiple variance structures and multiple genetic parameters can be also performed, as well as comparison between different models and genomic BLUP analysis. The AFEchidna package is free, please email us ([email protected]) to get a copy if one is interested for it. The AFEchidna package is developed to expand free genetic assessment software with the expectation that its efficiency could be close to the commercial software.

scholarly journals Environmental effects and repeatability estimates for sperm production and semen quality of Holstein bulls

2013 ◽  
Vol 56 (1) ◽  
pp. 971-979
Author(s):  
I. Boujenane ◽  
K. Boussaq
Keyword(s):  
Environmental Effects ◽  
Fixed Effects ◽  
Mixed Model ◽  
Volume Concentration ◽  
Semen Quality ◽  
Random Effect ◽  
Sperm Production ◽  
Holstein Bulls ◽  
Effect Of Age

<p><strong>Abstract.</strong> The purpose of this study was to estimate environmental effects on semen production of artificial insemination (AI) Holstein bulls managed under Moroccan conditions. A total of 4<span class="thinspace"></span>046 ejaculates collected from 34 Holstein bulls in the years 2009&ndash;2013 were analysed. Studied variables were volume, concentration, total number of spermatozoa, mass motility, individual motility and post-thawing motility. Data were analysed by REML method using the mixed model including the random effect of bull and the fixed effects of age at collection, season of collection, year of collection, interval between two collections and ejaculate order. The effect of age of bulls was significant for all studied variables, except for mass motility. Spring and winter were the best seasons for sperm production and quality. Bulls’ ejaculates collected once a day at one day interval produced monthly 30<span class="thinspace"></span>%, 86<span class="thinspace"></span>%, 156<span class="thinspace"></span>%, 183<span class="thinspace"></span>% and 185<span class="thinspace"></span>% more motile spermatozoa than those collected once a day at 2, 3, 4, 5 and 6 days interval, respectively. Moreover, bulls’ sperm collected twice a day at 1, 2, 3, 4, 5 and 6 days interval produced monthly (sum of motile spermatozoa obtained at 1st and 2nd ejaculates) 77<span class="thinspace"></span>%, 70<span class="thinspace"></span>%, 65<span class="thinspace"></span>%, 68<span class="thinspace"></span>%, 84<span class="thinspace"></span>% and 91<span class="thinspace"></span>% more than those collected once a day at the same interval, respectively. Repeatability estimates for semen traits were medium to high. They varied from 0.157 for mass motility to 0.411 for ejaculate volume. It was concluded that environmental factors clearly contribute to semen production in Holstein bulls and short intervals between collections and two collections per day are maximising sperm production.</p>

Sign in / Sign up

Export Citation Format

Share Document