2017
DOI: 10.1080/00949655.2017.1318876
|View full text |Cite
|
Sign up to set email alerts
|

Flexible Tweedie regression models for continuous data

Abstract: Tweedie regression models provide a flexible family of distributions to deal with non-negative highly right-skewed data as well as symmetric and heavy tailed data and can handle continuous data with probability mass at zero. The estimation and inference of Tweedie regression models based on the maximum likelihood method are challenged by the presence of an infinity sum in the probability function and non-trivial restrictions on the power parameter space. In this paper, we propose two approaches for fitting Twe… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
30
0

Year Published

2017
2017
2022
2022

Publication Types

Select...
9

Relationship

3
6

Authors

Journals

citations
Cited by 54 publications
(31 citation statements)
references
References 34 publications
1
30
0
Order By: Relevance
“…Discussion of the efficiency of the estimating function estimators is difficult due to the lack of a closed form for the Fisher information matrix. Bonat and Kokonendji (2016) showed in the context of Tweedie regression models that the quasi-score function provides asymptotically efficient estimators for the regression parameters, thus a similar result is expected for the Poisson-Tweedie regression model. Concerning the dispersion and power parameters, the fact that the sensitivity and variability matrices do not coincide indicates that the Pearson estimating functions are not optimum.…”
Section: Discussionsupporting
confidence: 59%
“…Discussion of the efficiency of the estimating function estimators is difficult due to the lack of a closed form for the Fisher information matrix. Bonat and Kokonendji (2016) showed in the context of Tweedie regression models that the quasi-score function provides asymptotically efficient estimators for the regression parameters, thus a similar result is expected for the Poisson-Tweedie regression model. Concerning the dispersion and power parameters, the fact that the sensitivity and variability matrices do not coincide indicates that the Pearson estimating functions are not optimum.…”
Section: Discussionsupporting
confidence: 59%
“…Of the fluency (WE, RE, SN), and network variables (, NC ), only M was best fit with a normal distribution. The tweedie variance function 109 was used to fit WE, RE, SN and , while the poisson-tweedie variance function 110 was used to fit the count variables, NC and Error.…”
Section: Resultsmentioning
confidence: 99%
“…To evaluate if the diet shift was systematic, that is occurred each spring preceding a mast, we used a similar approach with the RQ i , investigating potential inter‐annual effects on the relative quantity of each food in the diet of the population (RQ i , model 3). The Tweedie distribution is adapted to treat data with a right‐skewed distribution and high proportions of exact 0s (Bonat & Kokonendji, ). We included beech mast (Yes/No, models 1 and 2) or years (model 3).…”
Section: Methodsmentioning
confidence: 99%