Robit Regression: A Simple Robust Alternative to Logistic and Probit Regression

By Chuanhai Liu. 2005

154 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/f2e9d42f87b7651f06025ee07807cfeb08addf18

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Identifiability of Normal and Normal Mixture Models with Nonignorable Missing Data (2015, methodology): “s of freedom , then it satises Condition A when is known, and thus all parameters are identiable. The model with F() = T () for the missing mechanism is sometimes referred to as the Robit model (Liu, 2004). By varying the degrees of freedom , the Robit models can be used to approximate many missing ”
Parametric links for binary choice models: A Fisherian-Bayesian colloquy (2009, methodology): “ Liu (2004) . When is near zero, say less than 0.2, evaluation of the likelihood becomes problematic due to the dramatically heavy tails of the distribution.”
On the Conditional Distribution of the Multivariate t Distribution (2016, methodology): “…property, theMVTdistribution iswidely applied in robust data analysis including linear and nonlinear regressions (Lange, Little, and Taylor 1989; Liu 1994, 2004; Gelman et al. 2014), linear mixed effects models (Pinheiro, Liu, and Wu 2001), and sample selection models (Marchenko and Genton…”
Discussion of "Analysis of variance--why it is more important than ever" by A. Gelman (2005, methodology): “ed using this Gibbs sampler with Metropolis jumping for the nonconjugate conditional densities [see, e.g., Gelman, Carlin, Stern and Rubin (1995)] or data augmentation [see Albert and Chib (1993) and Liu (2002)]. In either case, once the simulations have approximately converged and posterior simulat”
On the Use of Cauchy Prior Distributions for Bayesian Logistic Regression (2015, background): “els, while the inverse cumulative distribution function (CDF) of the logistic distribution yields the logit link function, the inverse CDF of the Student-tdistribution yields the robit link function. Liu (2004) showed that the logistic link can be well approximated by a robit link with 7 degrees of ”
Demystifying Double Robustness: A Comparison of Alternative Strategies for Estimating a Population Mean from Incomplete Data (2007, background): “The logit link is well approximated by the robit with v ? 1, and smaller values of v lead to estimates that are more robust (Liu, 2004).”
A weakly informative default prior distribution for logistic and other regression models (2008, methodology): “We are motivated to consider the t family because flat-tailed distributions allow for robust inference [see, Berger and Berliner (1986), Lange, Little, and Taylor (1989)], and, as we shall see in Section 3, it allows easy and stable computation in logistic regression by placing iteratively weighted…”
The BUGS project: Evolution, critique and future directions (2009, background): “Numerous applications of the software can be found in the literature, in a wide array of application areas, including disease mapping [4], pharmacometrics [5], ecology [6], health-economics [7], genetics [8], archaeology [9], psychometrics [10], coastal engineering [11], educational performance [12]”
Flexible generalized t-link models for binary response data (2008, methodology): “In the multigrid Monte Carlo adjustment step, we use the multigrid Monte Carlo scheme of Liu & Sabatti (2000). First, we integrate outX ftom thejoint posterior distribution of (/3, 8, VI, w, X, z).”