Empirical comparison of the various spatial prediction models

In spatial econometrics, spatial statistics, and semiparametric statistics

Hajime Seya, Morito Tsutsumi, Yasushi Yoshida, Yuichiro Kawaguchi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This study compares the performance of various spatial prediction models which consider spatial dependence among the real estate data in spatial econometrics, spatial statistics, and semiparametric statistics. Thus far, surprisingly few researches have been conducted from this perspective. This study employs the dataset of apartment rent in the Tokyo 23 wards for empirical comparison. This study in particular focuses on a geoadditive model which considers both spatial dependence and nonlinearity of a hedonic function, and suggests the predictive performance of this method superior to those of the traditional methods like kriging or EM type algorithm.

Original languageEnglish
Title of host publicationProcedia - Social and Behavioral Sciences
Pages120-129
Number of pages10
Volume21
DOIs
Publication statusPublished - 2011
EventInternational Conference on Spatial Thinking and Geographic Information Sciences 2011, STGIS 2011 - Tokyo, Japan
Duration: 2011 Sep 142011 Sep 16

Other

OtherInternational Conference on Spatial Thinking and Geographic Information Sciences 2011, STGIS 2011
CountryJapan
CityTokyo
Period11/9/1411/9/16

Fingerprint

econometrics
statistics
Spatial Analysis
Pleasure
Tokyo
apartment
real estate
rent
performance
Research
Datasets

Keywords

  • Apartment rent
  • Geoadditive model
  • Nonlinearity
  • Spatial hedonic model
  • Spatial prediction

ASJC Scopus subject areas

  • Social Sciences(all)
  • Psychology(all)

Cite this

Empirical comparison of the various spatial prediction models : In spatial econometrics, spatial statistics, and semiparametric statistics. / Seya, Hajime; Tsutsumi, Morito; Yoshida, Yasushi; Kawaguchi, Yuichiro.

Procedia - Social and Behavioral Sciences. Vol. 21 2011. p. 120-129.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Seya, H, Tsutsumi, M, Yoshida, Y & Kawaguchi, Y 2011, Empirical comparison of the various spatial prediction models: In spatial econometrics, spatial statistics, and semiparametric statistics. in Procedia - Social and Behavioral Sciences. vol. 21, pp. 120-129, International Conference on Spatial Thinking and Geographic Information Sciences 2011, STGIS 2011, Tokyo, Japan, 11/9/14. https://doi.org/10.1016/j.sbspro.2011.07.025
Seya, Hajime ; Tsutsumi, Morito ; Yoshida, Yasushi ; Kawaguchi, Yuichiro. / Empirical comparison of the various spatial prediction models : In spatial econometrics, spatial statistics, and semiparametric statistics. Procedia - Social and Behavioral Sciences. Vol. 21 2011. pp. 120-129
@inproceedings{de7e7292641948ba9756cda0870731ba,
title = "Empirical comparison of the various spatial prediction models: In spatial econometrics, spatial statistics, and semiparametric statistics",
abstract = "This study compares the performance of various spatial prediction models which consider spatial dependence among the real estate data in spatial econometrics, spatial statistics, and semiparametric statistics. Thus far, surprisingly few researches have been conducted from this perspective. This study employs the dataset of apartment rent in the Tokyo 23 wards for empirical comparison. This study in particular focuses on a geoadditive model which considers both spatial dependence and nonlinearity of a hedonic function, and suggests the predictive performance of this method superior to those of the traditional methods like kriging or EM type algorithm.",
keywords = "Apartment rent, Geoadditive model, Nonlinearity, Spatial hedonic model, Spatial prediction",
author = "Hajime Seya and Morito Tsutsumi and Yasushi Yoshida and Yuichiro Kawaguchi",
year = "2011",
doi = "10.1016/j.sbspro.2011.07.025",
language = "English",
volume = "21",
pages = "120--129",
booktitle = "Procedia - Social and Behavioral Sciences",

}

TY - GEN

T1 - Empirical comparison of the various spatial prediction models

T2 - In spatial econometrics, spatial statistics, and semiparametric statistics

AU - Seya, Hajime

AU - Tsutsumi, Morito

AU - Yoshida, Yasushi

AU - Kawaguchi, Yuichiro

PY - 2011

Y1 - 2011

N2 - This study compares the performance of various spatial prediction models which consider spatial dependence among the real estate data in spatial econometrics, spatial statistics, and semiparametric statistics. Thus far, surprisingly few researches have been conducted from this perspective. This study employs the dataset of apartment rent in the Tokyo 23 wards for empirical comparison. This study in particular focuses on a geoadditive model which considers both spatial dependence and nonlinearity of a hedonic function, and suggests the predictive performance of this method superior to those of the traditional methods like kriging or EM type algorithm.

AB - This study compares the performance of various spatial prediction models which consider spatial dependence among the real estate data in spatial econometrics, spatial statistics, and semiparametric statistics. Thus far, surprisingly few researches have been conducted from this perspective. This study employs the dataset of apartment rent in the Tokyo 23 wards for empirical comparison. This study in particular focuses on a geoadditive model which considers both spatial dependence and nonlinearity of a hedonic function, and suggests the predictive performance of this method superior to those of the traditional methods like kriging or EM type algorithm.

KW - Apartment rent

KW - Geoadditive model

KW - Nonlinearity

KW - Spatial hedonic model

KW - Spatial prediction

UR - http://www.scopus.com/inward/record.url?scp=82555185699&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=82555185699&partnerID=8YFLogxK

U2 - 10.1016/j.sbspro.2011.07.025

DO - 10.1016/j.sbspro.2011.07.025

M3 - Conference contribution

VL - 21

SP - 120

EP - 129

BT - Procedia - Social and Behavioral Sciences

ER -