Tutorial on kernel estimation of continuous spatial and spatiotemporal relative risk

Kernel smoothing is a highly flexible and popular approach for estimation of probability density and intensity functions of continuous spatial data. In this role, it also forms an integral part of estimation of functionals such as the density-ratio or "relative risk" surface. Originally developed with the epidemiological motivation of examining fluctuations in disease risk based on samples of cases and controls collected over a given geographical region, such functions have also been successfully used across a diverse range of disciplines where a relative comparison of spatial density functions has been of interest. This versatility has demanded ongoing developments and improvements to the relevant methodology, including use spatially adaptive smoothers; tests of significantly elevated risk based on asymptotic theory; extension to the spatiotemporal domain; and novel computational methods for their evaluation. In this tutorial paper, we review the current methodology, including the most recent developments in estimation, computation, and inference. All techniques are implemented in the new software package sparr, publicly available for the R language, and we illustrate its use with a pair of epidemiological examples.

[1]  G. Terrell.  The Maximal Smoothing Principle in Density Estimation , 1990 .

[2]  Dai Feng,et al.  Computing and Displaying Isosurfaces in R , 2008 .

[3]  M. Wand.  Fast Computation of Multivariate Kernel Estimators , 1994 .

[4]  Tilman M. Davies,et al.  Spatial and social variables in the Bronze Age Phase 4 cemetery of Ban Non Wat, Northeast Thailand , 2015 .

[5]  Tilman M. Davies,et al.  Jointly optimal bandwidth selection for the planar kernel-smoothed density-ratio. , 2013, Spatial and spatio-temporal epidemiology.

[6]  J. Bithell.  An application of density estimation to geographical epidemiology. , 1990, Statistics in medicine.

[7]  J Belcher,et al.  Nonparametric methods. , 2001, Nurse researcher.

[8]  Rob J. Hyndman,et al.  A Bayesian approach to bandwidth selection for multivariate kernel density estimation , 2006, Comput. Stat. Data Anal..

[9]  M. Hazelton.  Testing for changes in spatial relative risk , 2017, Statistics in medicine.

[10]  Tilman M. Davies,et al.  Inference Based on Kernel Estimates of the Relative Risk Function in Geographical Epidemiology , 2009, Biometrical journal. Biometrische Zeitschrift.

[11]  Charles C. Taylor,et al.  Bootstrap choice of the smoothing parameter in kernel density estimation , 1989 .

[12]  L. Waller,et al.  Applied Spatial Statistics for Public Health Data: Waller/Applied Spatial Statistics , 2004 .

[13]  Martin L Hazelton,et al.  Generalizing the spatial relative risk function. , 2014, Spatial and spatio-temporal epidemiology.

[14]  D. W. Scott,et al.  Biased and Unbiased Cross-Validation in Density Estimation , 1987 .

[15]  Martin L. Hazelton,et al.  On the utility of asymptotic bandwidth selectors for spatially adaptive kernel density estimation , 2018, Statistics & Probability Letters.

[16]  P. Diggle,et al.  Kernel estimation of relative risk , 1995 .

[17]  M. C. Jones,et al.  On a class of kernel density estimate bandwidth selectors , 1991 .

[18]  Tilman M. Davies,et al.  Fast computation of spatially adaptive kernel estimates , 2017, Statistics and Computing.

[19]  S. Cornell,et al.  Dynamics of the 2001 UK Foot and Mouth Epidemic: Stochastic Dispersal in a Heterogeneous Landscape , 2001, Science.

[20]  M. C. Jones,et al.  Simple boundary correction for kernel density estimation , 1993 .

[21]  Martin L. Hazelton,et al.  Symmetric adaptive smoothing regimens for estimation of the spatial relative risk function , 2016, Comput. Stat. Data Anal..

[22]  J. S. Marron,et al.  Variable window width kernel estimates of probability densities , 1988 .

[23]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[24]  Virgilio Gómez-Rubio,et al.  Spatial Point Patterns: Methodology and Applications with R , 2016 .

[25]  Martin L. Hazelton,et al.  Boundary kernels for adaptive density estimators on regions with irregular boundaries , 2010, J. Multivar. Anal..

[26]  Ian Abramson.  On Bandwidth Variation in Kernel Estimates-A Square Root Law , 1982 .

[27]  M. C. Jones,et al.  A Brief Survey of Bandwidth Selection for Density Estimation , 1996 .

[28]  W. McMahon,et al.  Spatial Relative Risk Patterns of Autism Spectrum Disorders in Utah , 2014, Journal of Autism and Developmental Disorders.

[29]  J. Bithell.  Estimation of relative risk functions. , 1991, Statistics in medicine.

[30]  A B Lawson,et al.  Spatial statistical modeling of disease outbreaks with particular reference to the UK foot and mouth disease (FMD) epidemic of 2001. , 2005, Preventive veterinary medicine.

[31]  M. Hazelton,et al.  Cross‐validation Bandwidth Matrices for Multivariate Kernel Density Estimation , 2005 .

[32]  Adrian Baddeley,et al.  spatstat: An R Package for Analyzing Spatial Point Patterns , 2005 .

[33]  Andrew Bevan,et al.  Spatial methods for analysing large-scale artefact inventories , 2012, Antiquity.

[34]  Martin L. Hazelton,et al.  sparr: Analyzing Spatial Relative Risk Using Fixed and Adaptive Kernel Density Estimation in R , 2011 .

[35]  Peter Diggle,et al.  The geographical distribution of primary biliary cirrhosis in a well‐defined cohort , 2001, Hepatology.

[36]  Andrew B. Lawson,et al.  Statistical Methods in Spatial Epidemiology: Lawson/Statistical Methods in Spatial Epidemiology , 2006 .

[37]  Matthew P. Wand,et al.  Kernel Smoothing , 1995 .

[38]  M. C. Jones,et al.  A reliable data-based bandwidth selection method for kernel density estimation , 1991 .

[39]  Mark J. Brewer,et al.  A Bayesian model for local smoothing in kernel density estimation , 2000, Stat. Comput..

[40]  Shuowen Hu,et al.  Bayesian adaptive bandwidth kernel density estimation of irregular multivariate distributions , 2012, Comput. Stat. Data Anal..

[41]  E. Parzen.  On Estimation of a Probability Density Function and Mode , 1962 .

[42]  J. Marron,et al.  Smoothed cross-validation , 1992 .

[43]  Peter J. Diggle,et al.  SPLANCS: spatial point pattern analysis code in S-Plus , 1993 .

[44]  P. Diggle,et al.  Non-parametric estimation of spatial variation in relative risk. , 1995, Statistics in medicine.

[45]  Tilman M. Davies,et al.  Adaptive kernel estimation of spatial relative risk , 2010, Statistics in medicine.

[46]  M. C. Jones,et al.  Comparison of Smoothing Parameterizations in Bivariate Kernel Density Estimation , 1993 .

[47]  A. Cuevas,et al.  A comparative study of several smoothing methods in density estimation , 1994 .

[48]  Modelling dichotomously marked muscle fibre configurations. , 2013, Statistics in medicine.

[49]  Kernel estimation of risk surfaces without the need for edge correction , 2008, Statistics in medicine.

[50]  P. Diggle.  A Kernel Method for Smoothing Point Process Data , 1985 .

[51]  J. Faraway,et al.  Bootstrap choice of bandwidth for density estimation , 1990 .

[52]  A B Lawson,et al.  Applications of extraction mapping in environmental epidemiology. , 1993, Statistics in medicine.

[53]  D. W. Scott,et al.  Cross-Validation of Multivariate Densities , 1994 .

[54]  C. D. Kemp,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[55]  F. Campos,et al.  Spatial ecology of perceived predation risk and vigilance behavior in white-faced capuchins , 2014 .

[56]  Jeffrey S. Racine,et al.  Nonparametric Evaluation of Dynamic Disease Risk: A Spatio-Temporal Kernel Approach , 2011, PloS one.

[57]  L. Breiman,et al.  Variable Kernel Estimates of Multivariate Densities , 1977 .