Recursive Portfolio Selection with Decision Trees
Wolfgang Härdle () and
No SFB649DP2008-009, SFB 649 Discussion Papers from Humboldt University, Collaborative Research Center 649
A great proportion of stock dynamics can be explained using publicly available information. The relationship between dynamics and public information may be of nonlinear character. In this paper we offer an approach to stock picking by employing so-called decision trees and applying them to XETRA DAX stocks. Using a set of fundamental and technical variables, stocks are classified into three groups according to the proposed position: long, short or neutral. More precisely, by assessing the current state of a company, which is represented by fundamental variables and current market situation, well reflected by technical variables, it is possible to suggest if the current market value of a company is underestimated, overestimated or the stock is fairly priced. The performance of the model over the observed period suggests that XETRA DAX stock returns can adequately be predicted by publicly available economic data. Another conclusion of this study is that the implied volatility variable, when included into the training sample, boosts the predictive power of the model significantly.
Keywords: CART; decision trees in finance; nonlinear decision rules; asset management; portfolio optimisation (search for similar items in EconPapers)
JEL-codes: C14 C49 G11 G12 (search for similar items in EconPapers)
Pages: 27 pages
References: View references in EconPapers View complete reference list from CitEc
Citations: Track citations by RSS feed
Downloads: (external link)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:hum:wpaper:sfb649dp2008-009
Access Statistics for this paper
More papers in SFB 649 Discussion Papers from Humboldt University, Collaborative Research Center 649 Contact information at EDIRC.
Bibliographic data for series maintained by RDC-Team ().