Lognormal and Gamma Mixed Negative Binomial Regression
نویسندگان
چکیده
In regression analysis of counts, a lack of simple and efficient algorithms for posterior computation has made Bayesian approaches appear unattractive and thus underdeveloped. We propose a lognormal and gamma mixed negative binomial (NB) regression model for counts, and present efficient closed-form Bayesian inference; unlike conventional Poisson models, the proposed approach has two free parameters to include two different kinds of random effects, and allows the incorporation of prior information, such as sparsity in the regression coefficients. By placing a gamma distribution prior on the NB dispersion parameter r, and connecting a lognormal distribution prior with the logit of the NB probability parameter p, efficient Gibbs sampling and variational Bayes inference are both developed. The closed-form updates are obtained by exploiting conditional conjugacy via both a compound Poisson representation and a Polya-Gamma distribution based data augmentation approach. The proposed Bayesian inference can be implemented routinely, while being easily generalizable to more complex settings involving multivariate dependence structures. The algorithms are illustrated using real examples.
منابع مشابه
Lognormal and Gamma Mixed Negative Binomial Regression : Supplementary Material
Proof: The first and second derivatives of f(z) = − ln(1 − pz) are f ′(z) = p 1−pz and f ′′(z) = p (1−pz)2 = (f ′(z))2, respectively, and f j(0) = 0 for j ≥ 1, therefore d dzm f j(z) ∣∣ z=0 = 0 for j > m and thus (2) is true for j > m. When n = j = 1, we have d n dzn f j(z) ∣∣ z=0 = f ′ k(0) = p, ∑n j′=1 F (n, j ′) j! (j−j′!) [f ′(z)]nf j−j ′ (z) ∣∣∣ z=0 = p, and F (n, j)j!pn = p, therefore, (2...
متن کاملEstimation of Count Data using Bivariate Negative Binomial Regression Models
Abstract Negative binomial regression model (NBR) is a popular approach for modeling overdispersed count data with covariates. Several parameterizations have been performed for NBR, and the two well-known models, negative binomial-1 regression model (NBR-1) and negative binomial-2 regression model (NBR-2), have been applied. Another parameterization of NBR is negative binomial-P regression mode...
متن کاملComparison of Confidence and Prediction Intervals
1 2 A major focus for transportation safety analysts is the development of crash prediction models, a 3 task for which an extremely wide selection of model types are available. Perhaps the most common 4 crash prediction model is the negative binomial (NB) regression model. The NB model gained 5 popularity due to its relative ease of implementation and its ability to handle overdispersion in 6 c...
متن کاملFunctional Form and Heterogeneity in Models for Count Data
This study presents several extensions of the most familiar models for count data, the Poisson and negative binomial models. We develop an encompassing model for two well-known variants of the negative binomial model (the NB1 and NB2 forms). We then analyze some alternative approaches to the standard log gamma model for introducing heterogeneity into the loglinear conditional means for these mo...
متن کاملRegression for citation data: An evaluation of different methods
Citations are increasingly used for research evaluations. It is therefore important to identify factors affecting citation scores that are unrelated to scholarly quality or usefulness so that these can be taken into account. Regression is the most powerful statistical technique to identify these factors and hence it is important to identify the best regression strategy for citation data. Citati...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- Proceedings of the ... International Conference on Machine Learning. International Conference on Machine Learning
دوره 2012 شماره
صفحات -
تاریخ انتشار 2012