淡江大學機構典藏:Item 987654321/114036
English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 62830/95882 (66%)
造访人次 : 4134395      在线人数 : 679
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜寻范围 查询小技巧：
  • 您可在西文检索词汇前后加上"双引号"，以获取较精准的检索结果
  • 若欲以作者姓名搜寻，建议至进阶搜寻限定作者字段，可获得较完整数据
    •  进阶搜寻


    jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/114036


    题名: ETF價格波動預測能力之探討
    其它题名: Investigating the forecast performance of volatility for exchange traded funds
    作者: 江宗軒;Chiang, Tsung-Hsuan
    贡献者: 淡江大學財務金融學系碩士班
    李沃牆;林惠娜;Lee, Wo-Chiang;Lin, Hui-Na
    关键词: GARCH;GJR-GARCH;EGARCH;隨機漫步;Diebold-Mariano Test;ETF;Random walk
    日期: 2017
    上传时间: 2018-08-03 14:39:35 (UTC+8)
    摘要: 近幾年ETF在台灣迅速地發展，隨著衍生性商品的多樣化，使得預測商品價格的波動變得格外重要且也日漸受到投資人的重視，因此本文欲透過RW、GARCH、EGARCH、GJR-GARCH做為預測波動模型去捕捉6檔兩岸成分ETF的波動性，再透過二種損失函數(loss function)與DM檢定去區分模型間的預測能力並進行優劣排序，最後找出相對最佳預測模型以提供市場參與者作為投資交易策略之重要參考依據。
    利用GARCH模型來描述ETF波動叢聚、厚尾等特性，其次，利用EGARCH和GJR-GARCH模型來捕捉波動具有不對稱的特性，最終將模型進行配適及預測。實證發現最佳模型的決定對於不同損失函數的選取是敏感的，相對來說EGARCH模型對於台灣ETF價格有較好的預測能力，顯示ETF價格波動呈非對稱性且具槓桿效果，即壞消息比好消息更容易引起市場較大幅度的波動，而中國ETF價格較無法藉由單一模型去描述及預測。此外，分別以每日真實波動性和PK變幅日波動性當做真實波動之代理變數，結果發現無法透過PK變幅日波動性去觀察其預測績效，因此，本文認為預測兩岸成分ETF價格波動性時，採用每日真實波動性當做波動代理變數會比PK變幅日波動性更為理想。
    In the recent years, ETFs developed in Taiwan rapidly. With the diversities of the Financial Derivatives, it is virtually important that forecasting the volatility of merchandise price, and it also be valued by investers. As a result, this paper want to use RW, GARCH, EGARCH, and GJR-GARCH for forecasting volatility model to catch the volatilities of six ETFs composing by Taiwan and China. Forthermore, using two kinds of loss functions and DM test to distinguish forecasting abilities between these models, and progress order of merit. Finally, finding the best forecasting model can give market participants important reference for investment trading strategy.
    The first, using GARCH model demonstrates characteristics of ETFs about volatility clustering and fat-tail.The next, using EGARCH and GJR-GARCH models catches volatilities with asymmetric characteristics. In the third place, find this paper fits and forecasts from these models. Empirical demonstrates that decision of the best model is sensitive for choosing different loss functions. Relatively speaking, EGARCH model has a much better forecasting ability for ETF price in Taiwan and demonstrates that pricing volatilities of ETF have asymmetric and levering effect. In other words, bad news can easily cause greater volatilities than good news in financial markets. But the price of ETF in China can not use the only model to demonstrate and forecast. In addition, this paper views real daily volatility and PK-range as the proxy of real volatilities. The results find that using PK-range is impossible to observe forecasting performance. As a result, this paper believe that using daily real volatility as volatility proxy is more ideal than PK-range when forecasting the pricing volatilities of ETF composing by Taiwan and China.
    显示于类别:[財務金融學系暨研究所] 學位論文

    文件中的档案:

    档案 描述 大小格式浏览次数
    index.html0KbHTML120检视/开启

    在機構典藏中所有的数据项都受到原著作权保护.

    TAIR相关文章

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