運用 Probit、Logit與 Tobit 模型預測連鎖牙醫診所患者自...
國立高雄大學高階法律暨管理碩士在職專班(EMLBA)

 

  • 運用 Probit、Logit與 Tobit 模型預測連鎖牙醫診所患者自費支出之行為模式 = An Investigation on the Expenditure Behaviorof Patients in Chain Dental Clinics by using Probit 、Logit and Tobit Models
  • 紀錄類型: 書目-語言資料,印刷品 : 單行本
    並列題名: An Investigation on the Expenditure Behaviorof Patients in Chain Dental Clinics by using Probit 、Logit and Tobit Models
    作者: 黃介良,
    其他團體作者: 國立高雄大學
    出版地: [高雄市]
    出版者: 撰者;
    出版年: [民99]2010
    面頁冊數: 105面圖,表 : 30公分;
    標題: Logit模型
    標題: Probit
    電子資源: http://handle.ncl.edu.tw/11296/ndltd/85467341792232417230
    摘要註: 本研究乃有鑑於健保財政日益困窘,政府無力擺脫政治因素另闢財源,只得將目標轉向醫療系統,嚴格控管健保金額的上漲。因此,降低健保收入、閞拓自費收入來源顯然成為診所之大勢所趨。過去關於自費醫療的研究,大多侷限於患者之滿意度調查、相關因素以及期望值之分析,是故,本研究以Logit、Probit及Tobit模型來解析自費患者的行為模式,希望能從計量經濟學的角度來推演出機率方程式,做為未來診所自費醫療經營管理的關鍵工具。本研究抽樣方式是以某特定牙醫聯盟之各個診所為單位,每間診所由19-69歲的患者中,各年齡層隨機抽出六位患者,總共抽出1695個樣本。最後實證結果得知:所在分院、年齡、指定醫師、初診主訴及學歷,相對於其他自變項有統計學上之顯著差異。三個模型中,Logit 模型之錯誤/準確比率為0.23,Probit模型之錯誤/準確比率為0.26,確實足以預測連鎖牙醫診所患者自費支出之行為模式,而Tobit 模型也提供了相當有用之預測自費金額之估計方程式。 The serious financial problem of National Health Insurance in Taiwan is getting difficult because of the political reason. Therefore, the way that government can do to improve the uprising medical insurance budget is to restrict the insurance payment limits. Meanwhile, private clinics will need to expand their income by increasing more self-payment services.Most of existing researches focused only on the investigation regarding factor analysis, as well as the expected value of patients’ satisfaction. This study hence tries to explore the spending behavior of dental patients by using Probit, Logit and Tobit econometric models. We hope to find certain expenditure patterns of dental patients from econometric methods to provide key information for clinic managers.This study uses the data from patients of specific chain dental clinics to randomly choose 6 observations in each age ranging from 19- to 69-year-old. Finally, we utilizes 1,695 observations to conclude that dental patients’ self-payment behaviors are significantly affected by regressors such as “Clinic,” “Age,” “Appointment” and “Chief complain.”The noise to signal ratio of Probit and Logit models are 0.23 and 0.26, respectively, which are acceptable in predicting dental patients’ self-payment probabilities. The Tobit regression also provides valuable information in forecasting the self-payment amount for each specific patient of the chain dental clinics.
館藏
  • 2 筆 • 頁數 1 •
 
310002026634 博碩士論文區(二樓) 不外借資料 學位論文 TH 008M/0019 349952 4483 2010 一般使用(Normal) 在架 0
310002026642 博碩士論文區(二樓) 不外借資料 學位論文 TH 008M/0019 349952 4483 2010 c.2 一般使用(Normal) 在架 0
  • 2 筆 • 頁數 1 •
評論
Export
取書館別
 
 
變更密碼
登入