Lucas(1976) 批評當時總體時間序列的計量分析方法,且主張傳統計量模型參數會隨體制及政策而改變,基於這些評論,於是許多對。嗜好(Taste)"及"技術"(Technology)" 結構參數估計的進論方法偭開始使用動態模型中的尤拉最適化條件(Euler Optimality Conditios)來進行估計。
然而,其中以Hansen(1982)所提出來的一般化動差估計法(Generalized Method of Moments)(簡稱GMM)最受矚目。此法乃源於一般化工具變數(GIVE),在不需強烈假設下進行估計。其估計過程大致可分為下列三個階段:
1.建立正交化條件え建立目標函數最小化2.過度確認限制(overidentifying restriction) 之檢定問題因其本身即涵蓋許多估計式,如GIVE,MLE,2SLS, 且能滿足有限樣本性質,快速數斂。此法目前已用於總體計量,非線性理性預期實證及財務金融計量上。而本文應用台灣總體時間序列於資產訂價模型的GMM參數估計過程,證明了資料的適用性。另外,蒙地卡羅(Monte Carlo) 實驗設計模擬亦應用在本文研究,來探討有限樣本下的統計量之行為,並獲致適當的推論。 / Lucas(1976) criticized the existing strategies for econometricic analysis of macroeconomic time series and argues that papameters of traditional econometric models are not invariant with respect to shifts in policy regimes. In response to that criticism, several inference strategies for "taste and technology" structural parameter models using Euler optimality conditions in dynamic models were suggested.
Hansen's(1982) Generalized Method of Moments(henceforth GMM) instrumental
variables procedure is among the most notable inference strrategies
for structural parameters.
The procedure of GMM may consist three steps: (l)Set-up of the orthogonality
conditions (2).Minimizing the objective function. (3)Test
of the overidentifying restrictions
In this paper we can understand the statistical properties of GMM
estimator of Consumption-Based structural parameters obtained from
Capital Asset Pricing Model by the use of Monte Carlo Simualtion .
Identifer | oai:union.ndltd.org:CHENGCHI/B2002004709 |
Creators | 李沃牆, LI, WO-QIANG |
Publisher | 國立政治大學 |
Source Sets | National Chengchi University Libraries |
Language | 中文 |
Detected Language | English |
Type | text |
Rights | Copyright © nccu library on behalf of the copyright holders |
Page generated in 0.015 seconds