Please use this identifier to cite or link to this item: https://ah.lib.nccu.edu.tw/handle/140.119/97604
題名: LISREL模式中分段估計結構參數之分析及涵意
作者: 劉子鍵 ; 蔡坤宏
關鍵詞: 分段 ; 估計 ; 參數 ; 結構
日期: 九月-1994
上傳時間: 4-六月-2016
摘要: 利用LISREL進行實證研究時,為避免因測量指標過多而產生不適當解,研究者常 會採用分段估計作為變通的方式。至於分段估計對於參數估計值有何影響,值得加以探討。 本文主要的目的在藉由模擬程序來探討整體估計以及各種分段估計方法(迴歸加權法、加總 法之等權估計、加總法之結構負荷量加權估計和最小平方法)下結構參數(上游潛在變項對 下游潛在變項的影響)的精確性和正確性。 本研究結果發現:結構參數在上述四種分段估計方式下皆不如整體估計來得精確,且這四種 分段估計皆有低估的傾向。其中又以加總型的估計法最可能不包括真實值。這些現象支持一 些類同的研究結果。
To find out the influence on the structural parameter estimation using some alternative two-stage procedures instead of LISREL, we used Monte Carlo Simulation to generate pseudo data under a specified model. In the investigation we found that the performance of the regression weighted method and least square method are better than those of summation methods. However, these methods are prone to underestimate the parameters. For emperical study, these findings imply that we should be mindful of the bias in substantative interpretation when we use two-stage methods.
關聯: 教育與心理研究, 17,45-60
Journal of Education & Psychology
資料類型: article
Appears in Collections:期刊論文

Files in This Item:
File Description SizeFormat
index.html115 BHTML2View/Open
index.html151 BHTML2View/Open
Show full item record

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.