A simple parallel algorithm for large-scale portfolio problems
Purpose - Although the mean-variance portfolio selection model has been investigated in the literature, the difficulty associated with the application of the model when dealing with large-scale problems is limited. The aim of this paper is to close the gap by using the quadratic risk (standard devia...
Uloženo v:
| Vydáno v: | The journal of risk finance Ročník 11; číslo 5; s. 481 - 495 |
|---|---|
| Hlavní autoři: | , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
London
Emerald Group Publishing Limited
09.11.2010
Emerald Emerald Group Publishing, Ltd |
| Témata: | |
| ISSN: | 1526-5943, 2331-2947 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Shrnutí: | Purpose - Although the mean-variance portfolio selection model has been investigated in the literature, the difficulty associated with the application of the model when dealing with large-scale problems is limited. The aim of this paper is to close the gap by using the quadratic risk (standard deviation risk) function and finite iteration technique to remove difficulties in quadratic programming.Design methodology approach - Using van de Panne' approach, this paper proposes a finite technique to optimize large-scale portfolio selection problem.Findings - The proposal of parallel algorithm structure to the model provides a clearer decision framework to significantly enhance the efficiency of the portfolio selection process.Originality value - The proposal of parallel algorithm structure to the mean-variance portfolio selection model provides a clearer decision framework to significantly enhance the efficiency of the portfolio selection process. An empirical example that illustrates the application and benefits of the method is provided. |
|---|---|
| Bibliografie: | SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 14 ObjectType-Article-1 ObjectType-Feature-2 content type line 23 |
| ISSN: | 1526-5943 2331-2947 |
| DOI: | 10.1108/15265941011092068 |