scholarly journals Why analyze germination experiments using Generalized Linear Models?

2018 ◽  
Vol 40 (3) ◽  
pp. 281-287 ◽  
Author(s):  
Fábio Janoni Carvalho ◽  
Denise Garcia de Santana ◽  
Lúcio Borges de Araújo
Keyword(s):  
Generalized Linear Models ◽  
Binomial Distribution ◽  
Goodness Of Fit ◽  
Linear Models ◽  
Information Criterion ◽  
Binomial Model ◽  
Inappropriate Use ◽  
Copaiba Oil ◽  
Germination Experiments ◽  
Combined Data

Abstract: We compared the goodness of fit and efficiency of models for germination. Generalized Linear Models (GLMs) were performed with a randomized component corresponding to the percentage of germination for a normal distribution or to the number of germinated seeds for a binomial distribution. Lower levels of Akaikes’s Information Criterion (AIC) and Bayesian Information Criterion (BIC) combined, data adherence to simulated envelopes of normal plots and corrected confidence intervals for the means guaranteed the binomial model a better fit, justifying the importance of GLMs with binomial distribution. Some authors criticize the inappropriate use of analysis of variance (ANOVA) for discrete data such as copaiba oil, but we noted that all model assumptions were met, even though the species had dormant seeds with irregular germination.

Summary goodness-of-fit statistics for binary generalized linear models with noncanonical link functions

2015 ◽  
Vol 58 (3) ◽  
pp. 674-690 ◽  
Author(s):  
Jana D. Canary ◽  
Leigh Blizzard ◽  
Ronald P. Barry ◽  
David W. Hosmer ◽  
Stephen J. Quinn
Keyword(s):  
Generalized Linear Models ◽  
Goodness Of Fit ◽  
Linear Models ◽  
Fit Statistics ◽  
Link Functions

Large-scale model selection in misspecified generalized linear models

Biometrika ◽  
2021 ◽  
Author(s):  
Emre Demirkaya ◽  
Yang Feng ◽  
Pallavi Basu ◽  
Jinchi Lv
Keyword(s):  
Model Selection ◽  
Generalized Linear Models ◽  
Large Scale ◽  
Linear Models ◽  
Information Criterion ◽  
Scale Model ◽  
High Dimensional ◽  
Model Selection Consistency ◽  
New Information ◽  
Large Scale Model

Summary Model selection is crucial both to high-dimensional learning and to inference for contemporary big data applications in pinpointing the best set of covariates among a sequence of candidate interpretable models. Most existing work assumes implicitly that the models are correctly specified or have fixed dimensionality, yet both are prevalent in practice. In this paper, we exploit the framework of model selection principles under the misspecified generalized linear models presented in Lv and Liu (2014) and investigate the asymptotic expansion of the posterior model probability in the setting of high-dimensional misspecified models.With a natural choice of prior probabilities that encourages interpretability and incorporates the Kullback–Leibler divergence, we suggest the high-dimensional generalized Bayesian information criterion with prior probability for large-scale model selection with misspecification. Our new information criterion characterizes the impacts of both model misspecification and high dimensionality on model selection. We further establish the consistency of covariance contrast matrix estimation and the model selection consistency of the new information criterion in ultra-high dimensions under some mild regularity conditions. The numerical studies demonstrate that our new method enjoys improved model selection consistency compared to its main competitors.

scholarly journals Evaluation of non-linear models to describe development and fertility of codling moth at constant temperatures

2017 ◽  
Vol 20 (1) ◽  
pp. 3 ◽  
Author(s):  
H. Ranjbar Aghdam ◽  
Y. Fathipour ◽  
D. C. Kontodimas
Keyword(s):  
Goodness Of Fit ◽  
Linear Models ◽  
Information Criterion ◽  
Developmental Rate ◽  
Codling Moth ◽  
Coefficient Of Determination ◽  
Temperature Threshold ◽  
Optimum Temperature ◽  
Non Linear ◽  
Lower Temperature

Developmental rate of immature stages and age-specific fertility of females of codling moth at constant temperatures was modeled using non-linear models. The equations of Enkegaard, Analytis, and Bieri 1 and 2 were evaluated based on the value of adjusted R2 (R2adj) and Akaike information criterion (AIC) besides coefficient of determination (R2) and residual sum of squares (RSS). All models have goodness of fit to data especially for development [R2, R2adj, RSS and AIC ranged 0.9673-0.9917, 0.8601-0.9861, 0.08-6.7x10-4 and (-75.29) – (-46.26) respectively]. Optimum temperature (Topt) and upper threshold (Tmax) were calculated accurately (Topt and Tmax ranged 29.9-31.2oC and 35.9-36.7oC) by all models. Lower temperature threshold (Tmin) was calculated accurately by Bieri-1 model (9,9-10,8oC) whereas Analytis model (7,0-8,4oC) underestimated it. As far as fertility is concerned the respective values were better fitted near the optimum temperature (in 30oC) [R2 ,R2adj, RSS and AIC ranged 0,6966-0,7744, 0,5756-0,6455, 2,44-3,33 x10-4 and (-9,15)-7,15 respectively].

scholarly journals Comparison of the parameters of the lactation curve between normal and difficult calvings in Iranian Holstein cows

2019 ◽  
Vol 17 (1) ◽  
pp. e0401
Author(s):  
Navid Ghavi Hossein-Zadeh
Keyword(s):  
Dairy Cows ◽  
Milk Yield ◽  
Goodness Of Fit ◽  
Linear Models ◽  
Information Criterion ◽  
Holstein Cows ◽  
Coefficient Of Determination ◽  
Lactation Curve ◽  
Non Linear ◽  
Best Fit

To evaluate effect of dystocia on the lactation curve characteristics for milk yield and composition in Holstein cows, six non-linear models (Brody, Wood, Sikka, Nelder, Dijkstra and Rook) were fitted on 5,917,677 test day records for milk yield (MY), fat (FP) and protein (PP) percentages, fat to protein ratio (FPR) and somatic cell score (SCS) of 643,625 first lactation Holstein cows with normal calving or dystocia from 3146 herds which were collected by the Animal Breeding Center of Iran. The models were tested for goodness of fit using adjusted coefficient of determination, root means square error, Akaike’s information criterion and Bayesian information criterion. Rook model provided the best fit of the lactation curve for MY and SCS in normal and difficult calvers and dairy cows with dystocia for FP. Dijkstra model provided the best fit of the lactation curve for PP and FPR in normal and difficult calvers and dairy cows with normal calving for FP. Dairy cows with dystocia had generally lower 100-d, 200-d and 305-d cumulative milk yield compared with normal calvers. Time to the peak milk yield was observed later for difficult calvers (89 days in milk vs. 79 days in milk) with lower peak milk yield (31.45 kg vs. 31.88 kg) compared with normal calvers. Evaluation of the different non-linear models indicated that dystocia had important negative effects on milk yield and lactation curve characteristics in dairy cows and it should be reduced as much as possible in dairy herds.

scholarly journals Choice of non-linear models to determine the growth curve of meat-type quail

2021 ◽  
Vol 51 (2) ◽  
Author(s):  
Marta Jeidjane Borges Ribeiro ◽  
Fabyano Fonseca Silva ◽  
Maíse dos Santos Macário ◽  
José Aparecido Santos de Jesus ◽  
Claudson Oliveira Brito ◽  
...  
Keyword(s):  
Goodness Of Fit ◽  
Linear Models ◽  
Growth Curves ◽  
Information Criterion ◽  
Metabolizable Energy ◽  
Coefficient Of Determination ◽  
Parameter Estimates ◽  
Coturnix Coturnix ◽  
Non Linear ◽  
Meat Type

ABSTRACT: The objective of this study was to compare non-linear models fitted to the growth curves of quail to determine which model best describes their growth and check the similarity between models by analyzing parameter estimates.Weight and age data of meat-type European quail (Coturnix coturnix coturnix) of three lines were used, from an experiment in a 2 × 4 factorial arrangement in a completely randomized design, consisting of two metabolizable energy levels, four crude protein levels and six replicates. The non-linear Brody, Von Bertalanffy, Richards, Logistic and Gompertz models were used. To choose the best model, the Adjusted Coefficient of Determination, Convergence Rate, Residual Mean Square, Durbin-Watson Test, Akaike Information Criterion and Bayesian Information Criterion were applied as goodness-of-fit indicators. Cluster analysis was performed to check the similarity between models based on the mean parameter estimates. Among the studied models, Richards’ was the most suitable to describe the growth curves. The Logistic and Richards models were considered similar in the analysis with no distinction of lines as well as in the analyses of Lines 1, 2 and 3.

scholarly journals Modelling lactation curves for fat-to-protein ratio of milk in the first three lactations of Polish Holstein-Friesian cows

2021 ◽  
Vol 51 (3) ◽  
pp. 310-321
Author(s):  
A. Satola
Keyword(s):  
Correlation Coefficient ◽  
Fixed Effects ◽  
Goodness Of Fit ◽  
Linear Models ◽  
Random Effect ◽  
Information Criterion ◽  
Energy Status ◽  
Mathematical Functions ◽  
Milk Traits ◽  
Protein Ratio

Among milk traits, fat-to-protein ratio (FPR) is considered a potential measure of a cow’s energy status and is one of the selection criteria necessary to improve metabolic stability. Further genetic analyses require an appropriate model that describes the pattern of FPR changes throughout lactation. The objective of the study was to examine five mathematical functions to describe the lactation curve for FPR in the first three lactations of Polish Holstein-Friesians. The dataset contained FPR records for 5690 cows in the first lactation, 4081 cows in the second, and 2636 cows in the third lactation based on 48908, 34706, and 22097 test-day (TD) records, respectively. Using the MIXED procedure of SAS statistical analytics software, ten linear models (five with fixed effects only, and five with the additional random effect of cow) were fitted to the TD records. The goodness of fit was tested with Akaike's information criterion, residual variances and the correlation coefficient between the actual and estimated values. The model proposed by Ali and Schaeffer (1987) had the best fit to FPR in the first three parities, and the model of Wilmink (1987) provided the worst fit. The correlation coefficient between the actual and the estimated values of FPR was higher for models that included the random cow effect compared with models without this effect.

Sign in / Sign up

Export Citation Format

Share Document