Decision Technologies for Computational Finance pp 333-351 | Cite as
Quantification of Sector Allocation at the German Stock Market
Abstract
This paper reports upon a quantitative sector allocation of the German stock market, represented by the DAX (Deutscher Aktienindex). The purpose is to present an approach how to identify an appropriate sector allocation of the 10 DAX sectors quantitatively in advance to outperform the DAX. Fundamental and technical factors are used for the explanation of each of the 10 DAX sectors. With these models forecasts for each sector are given one month ahead. Two methodologies are used for forecasting the monthly returns: linear regression and neural networks. The next step is to take in addition to these forecasts the 24-month volatilities of the considered sectors and the corresponding correlation matrix to carry out a portfolio optimisation. With these received sector weights an out-of-sample performance comparison for the out-of sample range from January 1996 until June 1997 is conducted. Thus the focus of this study lies on two points: First, whether it is possible to outperform the benchmark DAX with a quantitative method. And second, whether nonlinear methodologies deliver added value to classical econometric methods.
Keywords
Interest Rate Sector Model Asset Allocation Monthly Return Producer Price IndexPreview
Unable to display preview. Download preview PDF.
Literature
- Beilner T., Mathes H. D., DTB DAX-Futures: Bewertung und Anwendung, Die Bank, 1990; 7: 388–395Google Scholar
- Bollerslev T., Generalized Autoregressive Conditional Heteroscedasticity, Journal of Econometrics, 1986; 31: 307–327CrossRefGoogle Scholar
- Burgess A. N., Modelling Asset Prices using the Portfolio of Cointegration Models Approach, London Business School, Department of Decision Science, Working Paper, 1997Google Scholar
- Clark P., Laxton D., Rose D., Asymmetry in the U.S. Output-Inflation Nexus, IMF Staff Papers, Vol. 43, No. 1, S. 216–251Google Scholar
- Dicks M., Output Gaps And All That, Lehman Brothers, June 1996Google Scholar
- Elton E. J., Gruber M. J., Modern Portfolio Theory and Investment Analysis, New York, 1991Google Scholar
- Engle R., Autoregressive Conditional Heteroscedasticity With Estimates of the Variance of U.K. Inflation, Econometrica, 1982; 50: 987–1007CrossRefGoogle Scholar
- Graf J., Zagorski P., Westphal M., Knöller S. „Strategische Asset Allocation mit Neuronalen Netzen: Prognosegesteuertes Portfoliomanagement”, in Quantitative Verfahren im Finanzmarktbereich, Schröder, M., ZEW-Wirtschaftsanalysen, Nomos, Band 5, 1996Google Scholar
- Hill T., Marquez M., O’Conner Remus, W., Artificial neural network models for forecasting and decision making, International Journal of Forecasting, 1994; 10: 5–15CrossRefGoogle Scholar
- Illmanen, Forecasting U.S. Bond Returns, Salomon Brothers Inc., United States Fixed-Income Research, Portfolio Strategies, August 1995Google Scholar
- Markowitz H. M., Portfolio Selection, Journal of Finance, 1952; 7: 67–91Google Scholar
- Moftakhar V., Varianzminimale Portefeuilles am deutschen Aktienmarkt, Beiträage zur Theorie der Finanzmärkte, Institut für Kapitalmarktforschung, J. W. GoetheUniversität, Frankfurt, Working Paper, 1994Google Scholar