Kernel Averaged Predictors for Spatio-Temporal Regression Models.

Matthew J Heaton, Alan E Gelfand
Author Information
  1. Matthew J Heaton: Department of Statistical Science, Duke University, Box 90251, Durham, NC 27708-0251 ( heaton@ucar.edu ; phone: 303-497-2884).

Abstract

In applications where covariates and responses are observed across space and time, a common goal is to quantify the effect of a change in the covariates on the response while adequately accounting for the spatio-temporal structure of the observations. The most common approach for building such a model is to confine the relationship between a covariate and response variable to a single spatio-temporal location. However, oftentimes the relationship between the response and predictors may extend across space and time. In other words, the response may be affected by levels of predictors in spatio-temporal proximity to the response location. Here, a flexible modeling framework is proposed to capture such spatial and temporal lagged effects between a predictor and a response. Specifically, kernel functions are used to weight a spatio-temporal covariate surface in a regression model for the response. The kernels are assumed to be parametric and non-stationary with the data informing the parameter values of the kernel. The methodology is illustrated on simulated data as well as a physical data set of ozone concentrations to be explained by temperature.

Keywords

References

  1. Biostatistics. 2000 Sep;1(3):279-92 [PMID: 12933509]
  2. J Am Stat Assoc. 2011 Mar;106(493):6-20 [PMID: 23459794]
  3. J Am Stat Assoc. 2007;102(480):1221-1234 [PMID: 19759840]
  4. J Am Stat Assoc. 2003;98(462):387-396 [PMID: 39421645]
  5. Epidemiology. 2000 May;11(3):320-6 [PMID: 10784251]
  6. Biometrics. 2009 Mar;65(1):282-91 [PMID: 18422792]
  7. Comput Stat Data Anal. 2009 Jun 15;53(8):2873-2884 [PMID: 20016667]
  8. Am J Epidemiol. 2005 Jul 1;162(1):80-8 [PMID: 15961590]

Grants

  1. R01 ES014843/NIEHS NIH HHS

Word Cloud

Created with Highcharts 10.0.0responsespatio-temporaldatacovariatesacrossspacetimecommonmodelrelationshipcovariatelocationpredictorsmaykernelapplicationsresponsesobservedgoalquantifyeffectchangeadequatelyaccountingstructureobservationsapproachbuildingconfinevariablesingleHoweveroftentimesextendwordsaffectedlevelsproximityflexiblemodelingframeworkproposedcapturespatialtemporallaggedeffectspredictorSpecificallyfunctionsusedweightsurfaceregressionkernelsassumedparametricnon-stationaryinformingparametervaluesmethodologyillustratedsimulatedwellphysicalsetozoneconcentrationsexplainedtemperatureKernelAveragedPredictorsSpatio-TemporalRegressionModelsDistributedlagGaussianprocessOzoneStochasticintegral

Similar Articles

Cited By (2)