Refine
Labor/Institute
Keywords
- Kreditmarkt (3)
- chaos (2)
- finance (2)
- nonlinear dynamics (2)
- recurrence analysis (2)
- Aktienrendite (1)
- Chaos (1)
- Chaostheorie (1)
- Citation Network (1)
- Finance (1)
Document Type
- Article (2)
- Working Paper (2)
- Conference Proceeding (1)
Language
- English (5)
In this study, we analyse the advantageous effects of neural networks in combination with wavelet functions on the performance of financial market predictions. We implement different approaches in multiple experiments and test their predictive abilities with different financial time series. We demonstrate experimentally that both wavelet neural networks and neural networks with data pre-processed by wavelets outperform classical network topologies. However, the precision of conducted forecasts implementing neural network algorithms still propose potential for further refinement and enhancement. Hence, we discuss our findings, comparisons with “buy-and-hold” strategies and ethical considerations critically and elaborate on future prospects.
Chaoticity Versus Stochasticity in Financial Markets: Are Daily S&P 500 Return Dynamics Chaotic?
(2021)
In this study, we present a combinatory chaos analysis of daily wavelet-filtered (denoised) S&P 500 returns (2000–2020) compared with respective surrogate datasets, Brownian motion returns and a Lorenz system realisation. We show that the dynamics of the S&P 500 return series consist of an almost equally divided combination of stochastic and deterministic chaos. The strange attractor of the S&P 500 return system is graphically displayed via Takens’ embedding and by spectral embedding in combination with Laplacian Eigenmaps. For the field of nonlinear and financial chaos research, we present a bibliometric analysis paired with citation network analysis. We critically discuss implications and future prospects.
This study provides an overview of the model evolution and research trends in the field of financial and risk modelling by applying a bibliometric approach from 2008–2019 and an overall citation network analysis. We present a content analysis of contributing authors, countries, journals, main topics, agreements, disagreements and frontiers within the research community and highlight quantitative features such as implemented models, aggregated model-family combinations and algorithms. Moreover, we describe the data sets employed by researchers. Finally, we discuss insights, such as the main statement, namely the non-existence of a “single-best”-approach as well as the future prospects of our findings.
Chaoticity Versus Stochasticity in Financial Markets: Are Daily S&P 500 Return Dynamics Chaotic?
(2022)
In this study, we empirically show the dynamics of daily wavelet-filtered (denoised) S&P 500 returns (2000–2020) to consist of an almost equally divided combination of stochastic and deterministic chaos, rendering the series unpredictable after expiration of the Lyapunov time, resulting in futile forecasting attempts. We achieve a clear distinction of the true nature of the underlying time series dynamics by applying a novel and combinatory chaos analysis framework comparing the wavelet-filtered S&P 500 returns with respective surrogate datasets, Brownian motion returns and a Lorenz system realisation. Furthermore, we are the first to show the strange attractor of especially the daily-frequented S&P 500 return system graphically via Takens´ embedding and by spectral embedding in combination with Laplacian Eigenmaps. Finally, we critically discuss implications and future prospects in terms of financial forecasting.