The Practical Implementation of Bayesian Model Selection

Chipman, Hugh A.; George, Edward I.; McCulloch, Robert E.

doi:10.1214/lnms/1215540964

Cited by 364 publications

(393 citation statements)

References 119 publications

Supporting

Mentioning

390

Contrasting

Order By: Relevance

“…This approach allowed us to weigh the evidence for each traffic term and see the amount of uncertainty in choosing the best model. The posterior model probabilities for each pollutant are shown by Equation (2) -Equation (4) (George and McCulloch 1997;Chipman et al 2001). …”

Section: Bayesian Variable Selection-withmentioning

confidence: 99%

See 1 more Smart Citation

Predicting residential indoor concentrations of nitrogen dioxide, fine particulate matter, and elemental carbon using questionnaire and geographic information system based data

Baxter

Clougherty

Paciorek³

et al. 2007

Atmospheric Environment

View full text Add to dashboard Cite

Previous studies have identified associations between traffic-related air pollution and adverse health effects. Most have used measurements from a few central ambient monitors and/or some measure of traffic as indicators of exposure, disregarding spatial variability and/or factors influencing personal exposure-ambient concentration relationships. This study seeks to utilize publicly available data (i.e., central site monitors, geographic information system (GIS), and property assessment data) and questionnaire responses to predict residential indoor concentrations of traffic-related air pollutants for lower socioeconomic status (SES) urban households.As part of a prospective birth cohort study in urban Boston, we collected indoor and outdoor 3-4 day samples of nitrogen dioxide (NO 2 ) and fine particulate matter (PM 2.5 ) in 43 low SES residences across multiple seasons from 2003 -2005. Elemental carbon concentrations were determined via reflectance analysis. Multiple traffic indicators were derived using Massachusetts Highway Department data and traffic counts collected outside sampling homes. Home characteristics and occupant behaviors were collected via a standardized questionnaire. Additional housing information was collected through property tax records, and ambient concentrations were collected from a centrally-located ambient monitor.The contributions of ambient concentrations, local traffic and indoor sources to indoor concentrations were quantified with regression analyses. PM 2.5 was influenced less by local traffic but had significant indoor sources, while EC was associated with traffic and NO 2 with both traffic and indoor sources. Comparing models based on covariate selection using p-values or a Bayesian approach yielded similar results, with traffic density within a 50m buffer of a home and distance from a truck route as important contributors to indoor levels of NO 2 and EC, respectively. The Bayesian approach also highlighted the uncertanity in the models. We conclude that by utilizing public databases and

show abstract

Section: Bayesian Variable Selection-withmentioning

confidence: 99%

“…We used c = n, making c large enough to acknowledge reasonable uncertainty in the effect estimates while still giving very unlikely effect estimates low prior probability. We also conducted sensitivity analysis by calculating the posterior probabilities with a range of c 's (5 -100) (Chipman et al 2001). …”

Section: Bayesian Variable Selection-withmentioning

confidence: 99%

Predicting residential indoor concentrations of nitrogen dioxide, fine particulate matter, and elemental carbon using questionnaire and geographic information system based data

Baxter

Clougherty

Paciorek³

et al. 2007

Atmospheric Environment

View full text Add to dashboard Cite

show abstract

“…A major practical advantage of the SSVS approach is that the researcher does not have to calculate the marginal likelihoods for each of the possible models. 3 …”

Section: Model Uncertaintymentioning

confidence: 99%

“…What this does is that at each iteration, only variables visited are used to simulate the alternative specific constant which will be used to average the CV estimate. This way the frequency that a variable is included in the model is used to weight the variable and follows the procedure proposed by Chipman, George and McCulloch (2001).…”

Section: As In Abidoye Et Al (2010) Let υ Smentioning

confidence: 99%

Model Uncertainty in Characterizing Recreation Demand

Abidoye

Herriges

2012

Environ Resource Econ

View full text Add to dashboard Cite

A Bayesian variable selection procedure is used to control for uncertainty in the specification of a recreational demand model. Specifically, we propose a model that draws on the Bayesian paradigm to integrate the variable selection process into the model and reflect the accompanying uncertainty about which is the "best" specification used for counterfactual predictions. The advantage of this procedure over previous non-Bayesian approaches is that overcomes the problem of pre-testing in specification searches. Keywords October 8, 2010Abstract A Bayesian variable selection procedure is used to control for uncertainty in the specification of a recreational demand model. In contrast to comparing models based on the likelihood values with unknown sampling properties (as in, e.g., Egan et al, 2009), we propose a model that draws on the Bayesian paradigm to integrate the variable selection process into the model and reflect the accompanying uncertainty about which is the "best" specification used for counterfactual predictions.

show abstract

“…The idea of Bayesian Model Averaging was set out by Leamer (1978), but has recently received a lot of attention in the statistics literature, including in particular Raftery, Madigan and Hoeting (1997), Hoeting, Madigan, Raftery and Volinsky (1999) and Chipman, George and McCulloch (2001). It has also been used in a number of econometric applications, including output growth forecasting (Min and Zellner (1993), Koop and Potter (2003)), cross-country growth regressions (Doppelhofer, Miller and Sala-i-Martin (2000) and Fernandez, Ley and Steel (2001)) and stock return prediction (Avramov (2002) and Cremers (2002) The models do not have to be linear regression models, but I shall henceforth assume that they are.…”

Section: Bayesian Model Averagingmentioning

confidence: 99%

Forecasting U.S. Inflation by Bayesian Model Averaging

Wright¹

2003

SSRN Journal

View full text Add to dashboard Cite

Recent empirical work has considered the prediction of inflation by combining the information in a large number of time series. One such method that has been found to give consistently good results consists of simple equal weighted averaging of the forecasts over a large number of different models, each of which is a linear regression model that relates inflation to a single predictor and a lagged dependent variable. In this paper, I consider using Bayesian Model Averaging for pseudo out-of-sample prediction of US inflation, and find that it gives more accurate forecasts than simple equal weighted averaging. This superior performance is consistent across subsamples and inflation measures. Meanwhile, both methods substantially outperform a naive time series benchmark of predicting inflation by an autoregression.

show abstract

The Practical Implementation of Bayesian Model Selection

Cited by 364 publications

References 119 publications

Predicting residential indoor concentrations of nitrogen dioxide, fine particulate matter, and elemental carbon using questionnaire and geographic information system based data

Predicting residential indoor concentrations of nitrogen dioxide, fine particulate matter, and elemental carbon using questionnaire and geographic information system based data

Model Uncertainty in Characterizing Recreation Demand

Forecasting U.S. Inflation by Bayesian Model Averaging

Contact Info

Product

Resources

About