| dc.contributor | 經濟學系 | |
| dc.creator (作者) | 林馨怡 | zh_TW |
| dc.date (日期) | 2013 | |
| dc.date.accessioned | 20-四月-2016 17:12:35 (UTC+8) | - |
| dc.date.available | 20-四月-2016 17:12:35 (UTC+8) | - |
| dc.date.issued (上傳時間) | 20-四月-2016 17:12:35 (UTC+8) | - |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/85805 | - |
| dc.description.abstract (摘要) | 本研究計劃預計解決分量迴歸在追蹤資料模型的內生性問題, 此模型可以分析解釋變數對 不同分量下的被解釋變數的異質性效果。 由於模型具有內生性問題, 本研究計劃將應用配 適值方法 (fitted value approach) 來解決內生性偏誤的問題。 因此, 本人將建構一個兩 階段的估計步驟, 其中第一個步驟為先對具內生性的解釋變數進行估計, 並得出其配適值, 接著第二步驟再利用此配適值來取代內生解釋變數, 然後進行追蹤資料分量迴歸之估計。 在第二階段估計時, 本研究計劃提出兩種估計方法, 預計分兩年完成。 第一年計劃將應用 Koenker (2004) 的具懲罰項的分量迴歸方法; 第二年計劃則將應用 Galvao (2011) 以及 Kato et al. (2012) 的分量迴歸固定效果模型估計方法。 由於估計方法不同, 估計式的大 樣本性質及其蒙地卡羅模擬設計也都不同, 因此未來兩年將分別完成不同估計方法下的大 樣本分析和小樣本模擬比較。 | |
| dc.description.abstract (摘要) | This proposal is concerned with estimating endogenous quantile regression (QR) model for panel data models, for revealing heterogeneity effects of regressors on the dependent variable. This proposal adopts the “fitted value” approach to eliminate the endogenous bias and develops a two-stage estimation procedure for the endoge- nous QR for panel data models. The first stage consists of estimating a fitted value for the endogenous variable. Under the assumption of independence between the instrumental variable and the disturbance terms of the endogenous QR for panel data models, the endogenous bias can be eliminated by replacing the endogenous variable with its fitted value. In this project, two estimation methods are consid- ered in the proposal. In the first year of the project, the penalized QR of Koenker (2004) is used and in the second year of the project, the QR fixed effect method of Galvao (2011) and Kato et al. (2012) is considered. The asymptotic properties of the estimators and the Monte Carlo simulations will be finished in the project. | |
| dc.format.extent | 856439 bytes | - |
| dc.format.mimetype | application/pdf | - |
| dc.relation (關聯) | 計畫編號 NSC 102-2410-H004-012 | |
| dc.subject (關鍵詞) | 內生性問題;配適值方法;分量迴歸;追蹤資料 | |
| dc.subject (關鍵詞) | Endogeneous;Fitted value approach;Quantile regression;Panel data | |
| dc.title (題名) | 追蹤資料的分量迴歸模型 | zh_TW |
| dc.title.alternative (其他題名) | The Endogenous Quantile Regression for Panel Data Models | |
| dc.type (資料類型) | report | |
