• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 2
  • 2
  • Tagged with
  • 2
  • 2
  • 2
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

逐次估計之研究

陳明山, CHEN, MING-SHAN Unknown Date (has links)
本文共一冊,約三萬五千餘字,全文分七章。第一章為緒論,說明研究動機與研究範 圍。第二章討論到底『逐次抽樣』比『固定樣本數抽樣』能獲益多少。第三章討論逐 次抉擇法則。統計學家逐次的抽取樣本,決定什麼時候停止抽樣;若停止抽樣,應採 取那一種終結行動。他的目標是使抽樣的費用加上決策損失的期望為最小。『貝氏』 和『大中取小』逐次抉擇法則是兩個較基本而重要的逐次抉擇法則,將分別在第四章 和第五章討論。第六章將探討逐次機率比檢定問題,它在逐次估計問題上是很基本而 重要的,本文將探討其一些基本而重要的特性。第七章為結論,說出本文研究結果, 並提出一些今後值得進一步研究的問題。
2

超越指數績效的投資組合最佳化模型 / Portfolio optimization models for enhanced index investment

朱志達, Chu, Chih Ta Unknown Date (has links)
建立指數基金時,通常是利用追蹤指數的技巧,選取少量的股票建構指數基金使得報酬率與標的指數(benchmark index)報酬率同步的投資組合。如果能建立包含少量股票的投資組合,就可達到指數追蹤的效果,那麼也能利用少量的股票建立績效可以超越指數基金的投資組合。本論文利用建構指數基金的方法以及大中取小的概念,挑選出一個績效可以超越標的指數的投資組合。本論文提出的模型亦考慮實務上交易所需的各項成本、整數交易單位與資產總類數等限制。因此,模型包含整數變數與二元變數。最後以台灣加權股價指數的相關資料做為實證研究的對象,實證結果顯示本論文提出的模型所建立的投資組合超越標的指數的績效平均年化報酬率25%。 / Setting up an index fund usually uses techniques of index-tracking that choosing few stocks forming a portfolio to obtain the same return rate as the benchmark index. Similarly we can use the same concept to set up a portfolio such that the performance is better than index’s. In this thesis we use index-tracking methods and minimax rule to obtain a portfolio which outperforms the benchmark index. In the proposed mathematical model we will consider the transaction costs, integer trading unit volume, and the total number of assets in the portfolio. Therefore the resulting model is a mixed integer nonlinear programming including integer variables and binary variables. Finally, the empirical study will be performed by using the data from the Taiwan stock market to verify the performance of our model. The empirical study shows that the portfolios created by our models outperform the benchmark index up to 25% in average.

Page generated in 0.0201 seconds