A novel generalized odd log-logistic Maxwell-based regression with application to microbiology

Abstract The presence of bimodality, heteroskedasticity, zero-inflation and nonlinear effects in covariables is common in several real data applications. In this context, new regressions are proposed for data with all these characteristics. The estimation of parameters follows the maximum likelihood method. For different fixed parameters, sample sizes and percentages of zeros, various simulations are performed to assess the behavior of the estimators. Quantile residuals are defined to evaluate the assumptions of the proposed regression. Its usefulness is illustrated by an experiment conducted to assess the soil microbiology in a sugarcane field.

[1]  R. Rigby,et al.  Generalized Additive Models for Location Scale and Shape (GAMLSS) in R , 2007 .

[2]  Emrah Altun,et al.  The generalized odd log-logistic family of distributions: properties, regression models and applications , 2017 .

[3]  B. Jørgensen.  Statistical Properties of the Generalized Inverse Gaussian Distribution , 1981 .

[4]  Anthony C. Atkinson,et al.  Plots, transformations, and regression : an introduction to graphical methods of diagnostic regression analysis , 1987 .

[5]  Christophe Mues,et al.  A zero-adjusted gamma model for mortgage loan loss given default , 2013 .

[6]  Zero-adjusted reparameterized Birnbaum–Saunders regression model , 2019, Statistics & Probability Letters.

[7]  J. S. Vasconcelos,et al.  A new regression model for bimodal data and applications in agriculture , 2020, Journal of applied statistics.

[8]  T. White.  Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics , 1990 .

[9]  Thiago G. Ramires,et al.  Estimating nonlinear effects in the presence of cure fraction using a semi-parametric regression model , 2018, Comput. Stat..

[10]  Fan Li,et al.  metamicrobiomeR: an R package for analysis of microbiome relative abundance data using zero-inflated beta GAMLSS and meta-analysis across studies using random effects models , 2019, BMC Bioinformatics.

[11]  G. Heller,et al.  Flexible Regression and Smoothing: Using Gamlss in R , 2017 .

[12]  Peter K. Dunn,et al.  Randomized Quantile Residuals , 1996 .

[13]  Gauss M. Cordeiro,et al.  A new generalized odd log-logistic flexible Weibull regression model with applications in repairable systems , 2018, Reliab. Eng. Syst. Saf..

[14]  S. Wood.  Generalized Additive Models: An Introduction with R , 2006 .

[15]  Zero-spiked regression models generated by gamma random variables with application in the resin oil production , 2018, Journal of Statistical Computation and Simulation.

[16]  M. Osaki,et al.  Analysis of bacterial communities on alkaline phosphatase genes in soil supplied with organic matter , 2008 .

[17]  W. Liesack,et al.  T-RFLP Analysis , 2004 .

[18]  T. Bruns,et al.  ITS primers with enhanced specificity for basidiomycetes ‐ application to the identification of mycorrhizae and rusts , 1993, Molecular ecology.

[19]  Vlasios Voudouris,et al.  Modelling skewness and kurtosis with the BCPE density in GAMLSS , 2012 .

[20]  Raydonal Ospina,et al.  A general class of zero-or-one inflated beta regression models , 2011, Comput. Stat. Data Anal..

[21]  F. D. Andreote,et al.  Microbial communities associated with plants: learning from nature to apply it in agriculture. , 2017, Current opinion in microbiology.

[22]  S van Buuren,et al.  Worm plot: a simple diagnostic device for modelling growth reference curves , 2001, Statistics in medicine.

[23]  G. Heller,et al.  The Zero-adjusted Inverse Gaussian distribution as a model for insurance claims. , 2006 .

[24]  H. Franco,et al.  Phosphate Sources and Filter Cake Amendment Affecting Sugarcane Yield and Soil Phosphorus Fractions , 2019, Revista Brasileira de Ciência do Solo.

[25]  A. Uitterlinden,et al.  Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA , 1993, Applied and environmental microbiology.