This paper studies the addition of linear constraints to Support Vector Regression when kernel is linear. Adding those into problem allows add prior knowledge on estimator obtained, such as finding positive vector, probability vector or monotone data. We prove that related optimization stays a semi-definite quadratic problem. also propose generalization Sequential Minimal Optimization algorithm...

We study generalized linear models for time series of counts, where serial dependence is introduced through a dependent latent process in the link function. Conditional on the covariates and the latent process, the observation is modelled by a negative binomial distribution. To estimate the regression coefficients, we maximize the pseudolikelihood that is based on a generalized linear model wit...

In many automatic speech recognition (ASR) applications, maximum likelihood linear regression (MLLR), and feature-based maximum likelihood linear regression (FMLLR) are used for speaker adaptation. This paper investigates a possible generalization of FMLLR which addresses the degradation in the performance of ASR systems due to small —possibly time-varying— perturbations of the training and the...

In this paper, we consider Bayesian estimation and model determination for two-way contingency tables. The analysis is based on partitioning the parameter space of a saturated log-linear model into parameter subspaces identiied by considering the natural symmetries of the data. Prior distributions may then be chosen to respect invariance restrictions required by the available prior information....

Zipf (1949) already noted that the linear relationship that he observed between log frequency and log rank is strongest in the middle range: both very high and very low frequency items tend to deviate from the log-log regression line. In this paper the causes for such deviations are investigated and a more detailed statistical model is offered. The subgeometric mean property of frequency counts...

We study the influence of explanatory variables in prediction by looking at the distribution of the log-odds ratio. We also consider the predictive influence of a subset of unobserved future variables on the distribution of log-odds ratio as well as in a logistic model, via the Bayesian predictive density of a future observation. This problem is considered for dichotomous, as well as continuous...

The paper considers general multiplicative models for complete and incomplete contingency tables that generalize log-linear and several other models and are entirely coordinate free. Sufficient conditions of the existence of maximum likelihood estimates under these models are given, and it is shown that the usual equivalence between multinomial and Poisson likelihoods holds if and only if an ov...

In regression models for categorical data a linear model is typically related to the response variables via a transformation of probabilities called the link function. We introduce an approach based on two link functions for binary data named log-mean (LM) and log-mean linear (LML), respectively. The choice of the link function plays a key role for the interpretation of the model, and our appro...

The importance of litter decomposition to carbon and nutrient cycling has motivated substantial research. Commonly, researchers fit a single-pool negative exponential model to data to estimate a decomposition rate (k). We review recent decomposition research, use data simulations, and analyze real data to show that this practice has several potential pitfalls. Specifically, two common decisions...