English  |  正體中文  |  简体中文  |  Items with full text/Total items : 62797/95867 (66%)
Visitors : 3746424      Online Users : 569
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/33902


    Title: 摺刀事件研究法之探討與應用
    Other Titles: Jackknife methods for event studies
    Authors: 詹凱婷;Tsan, Kai-ting
    Contributors: 淡江大學統計學系碩士班
    林志娟;Lin, Jyh-jiuan
    Keywords: 市場模型;期望報酬率;異常報酬率;檢定力;檢定大小;Market Model;Expected Returns;Abnormal Returns;power;Size.
    Date: 2009
    Issue Date: 2010-01-11 04:39:27 (UTC+8)
    Abstract: 本篇文章以台灣股票市場的日報酬率為研究對象,利用有母數方法、無母數方法及摺刀法,模擬樣本數在 25 及 50 的情況下檢定事件日是否會產生異常報酬,並以人為的方式在事件期增加不同水準異常報酬,分別探討各檢定方法在顯著水準 α 為 0.05 及 0.1 的檢定力。
    假設當事件發生時不會引起事件期異常報酬率的變異數改變的情況下,除了符號檢定法的檢定大小在小樣本底下低估發生型 I 誤的機率,而在大樣本底下則有過度拒絕虛無假設的情況外,其他檢定方法的檢定大小及檢定力表現都很好,其中以 Delete-one 摺刀法最佳;當事件發生時會引起事件期異常報酬率的變異數改變情況下, 以Christie 變異數增加模式及 Beaver 變異數增加模式模擬結果,不管在小樣本或大樣本及兩種變異數增加模式下都以 Delete-one 摺刀法檢定能力較穩健;另外比較 Christie 變異數增加模式及 Beaver 變異數增加模式的模擬結果,以 Beaver 變異數增加模式在增加不同水準異常報酬率底下的檢定力較 Christie 變異數增加模式佳。

    表單編號:ATRX-Q03-001-FM030-01
    The thesis aims to compare the performance of event study tests using daily return rates of Taiwan’s stock market. Under the number of simulated samples of 25 and 50 securities for each portfolio, respectively, parametric, non-parametric and jackknife methods are studied on testing whether the abnormal return is statistically significant on the event day. In order to investigate the power of the test methods under significant levels of 0.05 and 0.1, different levels of abnormal returns are artificially added. Assume that there is no change of the abnormal returns variance during the event period, most of the tests have certain power of detecting the abnormal return, especially, delete-one jackknife method outperforms the event study test methods in terms of size and power. Except it shows that sign test underestimates the occurrence of type I error probability (namely size) for small sample size and over rejects it for large sample size. Furthermore, two types of variance changes cases adopting Christie''s (1983) model and Beaver''s (1968) model are also studied. The simulation results also shows the delete-one jackknife method still outperforms the rest of the event study test methods and robust even under the variance changes cases in terms of the size and power.

    表單編號:ATRX-Q03-001-FM031-01
    Appears in Collections:[統計學系暨研究所] 學位論文

    Files in This Item:

    File SizeFormat
    0KbUnknown267View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback