Smoothly mixing regressions

By John Geweke, Michael P. Keane. 2007

190 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Finance, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/beb63318cfa3858ee4bf1ac1e6e1a23a820cab8c

How researchers see this paper

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

In Artificial Intelligence

In Finance

In Statistics and Probability

Citations in the corpus

Bayesian Time-Varying Quantile Forecasting for Value-at-Risk in Financial Markets (2011, methodology): “We also considered the MCMC estimated method of Geweke and Keane (2007) (GK), that employs a smooth mixture of Gaussian regression models. This method required a long pre-sample period of returns: we employed an extra 1000 daily returns, pre-2001, to employ this method and used all the same settings”
Stick-breaking autoregressive processes (2011, background): “…y, the infinite mixture model can be extended by allowing the kernel k to depend on x (Leslie et al., 2007; Chib and Greenberg, 2009), the mixing measure G to depend on x (De Iorio et al., 2004; Müller et al., 2004; Griffin and Steel, 2006; Dunson et al., 2007) or both (Geweke and Keane, 2007).”
POSTERIOR CONSISTENCY IN CONDITIONAL DENSITY ESTIMATION BY COVARIATE DEPENDENT MIXTURES (2011, methodology): “It has been suggested in the literature (Geweke and Keane (2007), Villani et al. (2009)) that covariate dependent mixture models in which locations also depend on covariates perform well in applications.”
When Gaussian Process Meets Big Data: A Review of Scalable GPs (2018, background): “where gi(x) is the gating function, which usually takes a parametric form such as the softmax or probit function [120], [121], and can be thought as the probability p(z = i) = πi that the expert indicator z is i , i.”
Regression Density Estimation With Variational Methods and Stochastic Approximation (2012, methodology): “…of 4646 daily returns from January 1, 1990, to May 29, 2008, and our response yt is log pt/pt−1, where pt is the closing S&P500 index on day t. Geweke and Keane (2007) and Li, Villani, and Kohn (2010b) considered time series models for the data using mixtures of regressions where the…”
Simultaneous Linear Quantile Regression: A Semiparametric Bayesian Approach (2012, methodology): “It is possible to avoid the monotonicity problem altogether by specifying a nonparametric model for the conditional distribution FY (y | x) and then inverting it to derive conditional quantile curves QY (τ | x) (Scaccia and Green 2003; Geweke and Keane 2007; Taddy and Kottas 2010).”
A simple nonparametric approach to estimating the distribution of random coefficients in structural models (2016, unknown): “We obtain the mixing ratios θ̂S as in (12) using the simulated basis functions and our distribution function estimator is still F̂N,S (β), which belongs to FM R .”
Grouped Patterns of Heterogeneity in Panel Data (2015, background): “(32) We start by bounding Zig(θ, α), for all (θ, α) ∈ Nη, by a quantity that does not depend on (θ, α).”
Bayesian Multivariate Time Series Methods for Empirical Macroeconomics (2010, unknown): “Even those who wish to do model comparison often use other metrics (e.g. Geweke and Keane, 2007, uses cross-validation).”
Learning to Average Predictively over Good and Bad: Comment on: Using Stacking to Average Bayesian Predictive Distributions (2018, background): “When the uncertainty level tends to zero then the mixture of experts or the smoothly mixing regressions model is recovered as limiting case, see Geweke and Keane(2007), Jacobs etal.(1991). Clearly, the analysis is also conditional upon the choice of the number n and the information specified in the ”
Mixtures of Experts Models (2018, methodology): “The component weights are usually modelled using a MNL model, but this need not be the case; Geweke & Keane (2007) of the distribution f g ( y i | θ g ( x i )) depends on the type of outcome data under study.”