Abstract Within the field of stock market prediction a controversial discussion between technicians and fundamentalists concerning the qualification of these different methods has taken place. On the one hand, experts use so-called charts to extract those formations they regard to be significant for the future development of stock prices. This procedure requires extensive experience in recognizing and interpreting the patterns and can also contain many sources of error. On the other hand, the fundamentalists have to decide which information, even regarding other influences, they consider. Therefore, it is intended to link both perspectives. Some analysts use statistical methods (i.e. moving averages or auto-regressive models) in order to indicate important clues concerning future trends in stock prices. The ARIMA-Model combines the abilities of these two methods. Another problem-solving approach uses Artificial Neural Networks (ANN). They are in a loose sense based on concepts derived from research into the nature of the brain [16]. Particularly the ANN's ability of filtering ‚noisy’ influences, which may be caused by differential behaviour of various investors seems to predetermine this approach. Our intention for both approaches is a short-term prediction (the following day's stock price). In spite of that this will be extended to a medium-term prediction (a monthly forecast).
[1]
Michèle Hibon,et al.
Accuracy of Forecasting: An Empirical Investigation
,
1979
.
[2]
David E. Rumelhart,et al.
Predicting the Future: a Connectionist Approach
,
1990,
Int. J. Neural Syst..
[3]
Neuronale Netze zur Entscheidungsunterstützung in der Betriebswirtschaft
,
1991
.
[4]
E. Fama.
EFFICIENT CAPITAL MARKETS: A REVIEW OF THEORY AND EMPIRICAL WORK*
,
1970
.
[5]
Christel Kemke,et al.
Der Neuere Konnektionismus
,
1988,
Inform. Spektrum.
[6]
Helmut Uhlir.
Überprüfung der Random-Walk-Hypothese auf dem österreichischen Aktienmarkt
,
1979
.
[7]
Berndt Müller,et al.
Neural networks: an introduction
,
1990
.
[8]
Bernd Brogsitter.
Ableitung optimaler Strategien am Aktienmarkt
,
1977
.
[9]
Walter Vandaele,et al.
Applied Time Series and Box-Jenkins Models
,
1983
.
[10]
Leon N. Cooper,et al.
An overview of neural networks: early models to real world systems
,
1990
.