Shrinkage estimators of the spatial relative risk function.
cross-validation
kernel smoothing
lasso
local likelihood
penalization
Journal
Statistics in medicine
ISSN: 1097-0258
Titre abrégé: Stat Med
Pays: England
ID NLM: 8215016
Informations de publication
Date de publication:
10 11 2023
10 11 2023
Historique:
revised:
26
06
2023
received:
30
01
2023
accepted:
01
08
2023
medline:
23
10
2023
pubmed:
21
8
2023
entrez:
20
8
2023
Statut:
ppublish
Résumé
The spatial relative risk function describes differences in the geographical distribution of two types of points, such as locations of cases and controls in an epidemiological study. It is defined as the ratio of the two underlying densities. Estimation of spatial relative risk is typically done using kernel estimates of these densities, but this procedure is often challenging in practice because of the high degree of spatial inhomogeneity in the distributions. This makes it difficult to obtain estimates of the relative risk that are stable in areas of sparse data while retaining necessary detail elsewhere, and consequently difficult to distinguish true risk hotspots from stochastic bumps in the risk function. We study shrinkage estimators of the spatial relative risk function to address these problems. In particular, we propose a new lasso-type estimator that shrinks a standard kernel estimator of the log-relative risk function towards zero. The shrinkage tuning parameter can be adjusted to help quantify the degree of evidence for the existence of risk hotspots, or selected to optimize a cross-validation criterion. The performance of the lasso estimator is encouraging both on a simulation study and on real-world examples.
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
4556-4569Informations de copyright
© 2023 The Author. Statistics in Medicine published by John Wiley & Sons Ltd.
Références
Bithell JF. An application of density estimation to geographical epidemiology. Stat Med. 1990;9:691-701.
Diggle P. A point process modelling approach to raised incidence of a rare phenomenon in the vicinity of a prespecified point. J R Stat Soc Ser A. 1990;153:349-362.
Diggle P, Rowlingson B. A conditional approach to point process modelling of elevated risk. J R Stat Soc: Ser A (Stat Soc). 1994;157(3):433-440.
Biggeri A, Barbone F, Lagazio C, Bovenzi M, Stanta G. Air Pollution and Lung Cancer in Trieste, Italy: Spatial Analysis of Risk as a function of distance from sources. Environ Health Perspect: Suppl. 1993;104(7):750.
Sabel CE, Gatrell AC, Löytönenc M, Maasiltad P, Jokelainene M. Modelling exposure opportunities: estimating relative risk for motor neurone disease in Finland. Soc Sci Med. 2000;50:1121-1137.
Zhang ZJ, Davies TM, Gao J, Wang Z, Jiang QW. Identification of high-risk regions for schistosomiasis in the Guichi region of China: an adaptive kernel density estimation-based approach. Parasitology. 2013;140(7):868-875.
Braeye T, De Schrijver K, Wollants E, Van Ranst M, Verhaegen J. A large community outbreak of gastroenteritis associated with consumption of drinking water contaminated by river water, Belgium, 2010. Epidemiol Infect. 2015;143(4):711-719.
Rosanowski S, Cogger N, Rogers C, Benschop J, Stevenson M. A description of the demographic characteristics of the New Zealand non-commercial horse population with data collected using a generalised random-tessellation stratified sampling design. Prev Vet Med. 2012;107(3-4):242-252.
Denzin N, Herwig V, Van Der Grinten E. Occurrence and geographical distribution of Canine Distemper Virus infection in red foxes (Vulpes vulpes) of Saxony-Anhalt, Germany. Vet Microbiol. 2013;162(1):214-218.
Wardrop N, Thomas L, Cook E, et al. The sero-epidemiology of Coxiella burnetii in humans and cattle, Western Kenya: evidence from a cross-sectional study. PLoS Negl Trop Dis. 2016;10(10):e0005032.
Miller M, Delroy-Buelles I, Bootland D, Lyon R. A spatial analysis of incident location and prehospital mortality for two United Kingdom Helicopter Emergency Medical Services (HEMS). Appl Spat Anal Policy. 2020;13(3):575-590.
Campos F, Fedigan L. Spatial ecology of perceived predation risk and vigilance behavior in white-faced capuchins. Behav Ecol. 2014;25(3):477-486.
Kelsall JE, Diggle PJ. Kernel estimation of relative risk. Bernoulli. 1995;1(1-2):3-16.
Kelsall JE, Diggle PJ. Non-parametric estimation of spatial variation in relative risk. Stat Med. 1995;14:2335-2343.
Davies TM, Hazelton ML. Adaptive kernel estimation of spatial relative risk. Stat Med. 2010;29:2423-2437.
Terrell GR. The maximal smoothing principle in density estimation. J Am Stat Assoc. 1990;85:470-477.
Davies T, Jones K, Hazelton M. Symmetric adaptive smoothing regimens for estimation of the spatial relative risk function. Comput Stat Data Anal. 2016;101:12-18.
Abramson I. On bandwidth variation in kernel estimation - a square root law. Ann Stat. 1982;10:1217-1223.
Davies T, Marshall J, Hazelton M. Tutorial on kernel estimation of continuous spatial and spatiotemporal relative risk. Stat Med. 2018;37(7):1191-1221.
Wand MP, Jones MC. Kernel Smoothing. London: Chapman & Hall; 1995.
Chacón J, Duong T. Multivariate Kernel Smoothing and Its Applications. London: Chapman and Hall/CRC; 2018.
Hazelton ML. Kernel estimation of risk surfaces without the need for edge correction. Stat Med. 2008;27:2269-2272.
Hazelton ML, Davies TM. Inference based on kernel estimates of the relative risk function in geographical epidemiology. Biometr J. 2009;51:98-109.
Diggle P. A kernel method for smoothing point process data. Appl Stat. 1985;34(2):138-147.
Scott DW. Multivariate Density Estimation: Theory, Practice and Visualization. New York: Wiley-Interscience; 1992.
Cronie O, Van Lieshout M. A non-model-based approach to bandwidth selection for kernel estimators of spatial intensity functions. Biometrika. 2018;105(2):455-462.
Tibshirani R, Hastie T. Local likelihood estimation. J Am Stat Assoc. 1987;82:559-567.
Fernando W, Ganesalingam S, Hazelton M. A comparison of estimators of the geographical relative risk function. J Stat Comput Simul. 2014;84(7):1471-1485.
Tibshirani R. Regression shrinkage and selection via the lasso. J R Stat Soc: Ser B (Methodol). 1996;58(1):267-288.
Hoerl A, Kennard R. Ridge regression: Biased estimation for nonorthogonal problems. Technometrics. 1970;12(1):55-67.
Baddeley A, Rubak E, Turner R. Spatial Point Patterns: Methodology and Applications With R. London: Chapman and Hall/CRC Press; 2015.
Prince MI, Chetwynd A, Diggle PJ, Jarner M, Metcalf JV, James OFW. The geographical distribution of primary biliary cirrhosis in a well-defined cohort. Hepatology. 2001;34:1083-1088.
Hazelton M. Testing for changes in spatial relative risk. Stat Med. 2017;36(17):2735-2749.
Choi E, Hall P. Nonparametric analysis of earthquake point-process data. In. Beachwood, Ohio: Institute of Mathematical Statistics; 2001:324-344.
R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2022.
Breiman L. Better subset regression using the nonnegative garrote. Technometrics. 1995;37(4):373-384.
Zhang Z, Chen D, Liu W, et al. Nonparametric evaluation of dynamic disease risk: A spatio-temporal kernel approach. PLOS One. 2011;6(3):e17381.