Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/151136
|
| Title: | 空間異質性檢測方法之比較及其在臺灣主要死因的應用
A Study of Spatial Heterogeneity Detection Methods and Their Application to Causes of Death in Taiwan |
| Authors: | 余清祥;梁穎誼;梁舒涵
Yue, Jack C.;Leong, Yin-yee;Liang, Shu-hen |
| Contributors: | 統計系 |
| Keywords: | 空間異質性;空間自相關;平均餘命;主要死因;探索性資料分析
Spatial heterogeneity;Spatial autocorrelation;Life expectancy;Leading causes of death;Explanatory data analysis |
| Date: | 2023-06 |
| Issue Date: | 2024-05-08 10:48:22 (UTC+8) |
| Abstract: | 臺灣土地面積雖然不大，各地居民的壽命差異卻相當大，民國110年臺北市、臺東縣男性的平均壽命差異高達8.9歲，隨著公共政策、社會資源的調整與挹注，預期國人健康及壽命的地區不平等將逐漸縮小。空間異質性可用於檢視健康與壽命的地區差異，而空間異質指的即是空間上的差異，相關檢定大致可分為三大類：總體檢定、區域檢定、焦點檢定，總體檢定可用於檢定全區域資料是否為空間同質，區域檢定多用於偵測高風險地區（或稱為群聚），焦點檢定則用於確認特定地區周圍是否有較高發生率。本文先使用電腦模擬比較三種異質性檢定中各一個具代表性的方法：Moran's I（總體檢定）、SaTScan（區域檢定）及Tango score test（焦點檢定），比較資料具有空間自相關及群聚時的偵測效果。其中模擬的實驗地區為二度空間，大小為5×5、7×7、9×9、⋯、21×21的格子點，檢測各方法在空間同質、空間自相關、群聚的效果。研究發現三種方法在空間同質性的結果大致相同，Moran's I在空間自相關時最敏感，而群聚存在時則以SaTScan效果最佳。本文套用上述研究結果至縣市、鄉鎮市區層級全死因及前三大主要死因（惡性腫瘤、心臟疾病、肺炎）。分析發現死亡率的空間分布具有時間一致性，但死亡率卻存有明顯的空間差異，高風險大多落在東南部山區。綜合而言，我國主要死因仍存有地區不平等的現象。
Taiwan is not a large country, but the discrepancy in life expectancy between areas is quite obvious. For example, in 2021, the difference in male life expectancy between Taipei City and Taitung County, a remote part of the southeast coast, is 9 years. We expect that the difference between regions will become smaller with the increasing reach of public resources. There is data through which this can be explored. Spatial heterogeneity can be used to explore regional inequality; there are three types of tests for checking spatial heterogeneity: global test, local test, and focused test. The global test can be used to test whether the data are spatially homogeneous; the local test is usually used to detect the location of high-risk areas (i.e., clusters); and the focused test can be used to confirm whether there are high incidence rates around a specific area. In this study, we select three common heterogeneity tests: Moran's I (global test), SaTScan (local test), and Tango score test (focused test), and evaluate whether these methods can detect spatial autocorrelation and/or clusters via simulation and empirical analysis. The simulation study is performed on a two-dimensional space, with lattice data of size 5 × 5, 7 × 7, 9 × 9, …, 21 × 21, under the assumption that the data should satisfy spatial homogeneity, spatial autocorrelation, and clustering. The empirical data considered are the overall township-level and major cause mortality rates in Taiwan. We found that these methods provide similar results in checking spatial homogeneity. Moran's I is the most sensitive test for spatial autocorrelation, and SaTScan is the best for testing the existence of clusters, followed by Tango score test. We also apply these methods to the mortality rates of the top three major causes of death (cancer, heart disease, and pneumonia) in Taiwan. We found that they change steadily, with the difference in mortality rates between two consecutive years satisfying spatially heterogeneity, while most clusters of mortality rates are located in the southeastern mountain areas. In general, there are still regional inequalities in the main causes of death. |
| Relation: | 人口學刊, Vol.66, pp.1-40 |
| Data Type: | article |
| DOI 連結: | https://doi.org/10.6191/JPS.202306_(66).0001 |
| DOI: | 10.6191/JPS.202306_(66).0001 |
| Appears in Collections: | [統計學系] 期刊論文
|
Files in This Item:
| File |
Description |
Size | Format | |
|---|
| index.html | | 0Kb | HTML | 93 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
