學術產出-期刊論文
| 題名 | Geographically weighted quantile regression for count Data |
| 作者 | 陳怡如 Chen, Vivian Yi-Ju;Wang, Shi-Ting |
| 貢獻者 | 統計系 |
| 關鍵詞 | Count data; Spatial nonstationarity; Quantile regression; Geographically weighted regression |
| 日期 | 2025-01 |
| 上傳時間 | 14-四月-2025 09:24:12 (UTC+8) |
| 摘要 | In recent years, a methodological framework known as geographically weighted quantile regression (GWQR) has emerged for spatial data analysis. This framework offers the abilities to simultaneously explore spatial heterogeneity or nonstationarity in regression relationships and to estimate various conditional quantile functions. However, the current configuration of GWQR is limited to the analysis of continuous dependent variables. Discrete count data are observed in many disciplines. Whenever modeling such outcomes is necessary, the conventional GWQR approach is inadequate and fails to provide comprehensive insights into count data. To address this issue, this study aims to extend the GWQR framework originally designed for continuous dependent variables to accommodate count outcomes. We introduce an approach called geographically weighted count quantile regression (GWCQR), wherein the model specification is based on the smoothing of count responses through a jittering procedure. A semiparametric counterpart that allows for the inclusion of both spatially varying and invariant coefficients is also discussed. Finally, the proposed techniques are applied to a dataset of dengue fever in Taiwan as an empirical illustration. |
| 關聯 | Statistics and Computing, Vol.35, article number: 33 |
| 資料類型 | article |
| DOI | https://doi.org/10.1007/s11222-025-10568-z |
| dc.contributor | 統計系 | |
| dc.creator (作者) | 陳怡如 | |
| dc.creator (作者) | Chen, Vivian Yi-Ju;Wang, Shi-Ting | |
| dc.date (日期) | 2025-01 | |
| dc.date.accessioned | 14-四月-2025 09:24:12 (UTC+8) | - |
| dc.date.available | 14-四月-2025 09:24:12 (UTC+8) | - |
| dc.date.issued (上傳時間) | 14-四月-2025 09:24:12 (UTC+8) | - |
| dc.identifier.uri (URI) | https://nccur.lib.nccu.edu.tw/handle/140.119/156523 | - |
| dc.description.abstract (摘要) | In recent years, a methodological framework known as geographically weighted quantile regression (GWQR) has emerged for spatial data analysis. This framework offers the abilities to simultaneously explore spatial heterogeneity or nonstationarity in regression relationships and to estimate various conditional quantile functions. However, the current configuration of GWQR is limited to the analysis of continuous dependent variables. Discrete count data are observed in many disciplines. Whenever modeling such outcomes is necessary, the conventional GWQR approach is inadequate and fails to provide comprehensive insights into count data. To address this issue, this study aims to extend the GWQR framework originally designed for continuous dependent variables to accommodate count outcomes. We introduce an approach called geographically weighted count quantile regression (GWCQR), wherein the model specification is based on the smoothing of count responses through a jittering procedure. A semiparametric counterpart that allows for the inclusion of both spatially varying and invariant coefficients is also discussed. Finally, the proposed techniques are applied to a dataset of dengue fever in Taiwan as an empirical illustration. | |
| dc.format.extent | 106 bytes | - |
| dc.format.mimetype | text/html | - |
| dc.relation (關聯) | Statistics and Computing, Vol.35, article number: 33 | |
| dc.subject (關鍵詞) | Count data; Spatial nonstationarity; Quantile regression; Geographically weighted regression | |
| dc.title (題名) | Geographically weighted quantile regression for count Data | |
| dc.type (資料類型) | article | |
| dc.identifier.doi (DOI) | 10.1007/s11222-025-10568-z | |
| dc.doi.uri (DOI) | https://doi.org/10.1007/s11222-025-10568-z |