Nonlinear and Nonparametric Methods for Analyzing Financial Time Series

Franke, J.

doi:10.1007/978-3-642-58409-1_27

Cited by 8 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another and frequently even more important problem is selecting the number and kind of exogenous variables serving as part of the inputs of the network. If one is willing to spend considerable time on model building and if the necessary data are available like in Franke [8,9] a further improvement is, therefore, to be expected compared to the examples above.…”

Section: Neural Network Based Risk Managementmentioning

confidence: 94%

“…Some case studies in Franke [8,9] show that using additional exogenous information represented by a high-dimensional X t helps in forecasting financial time series and in developping portfolio management strategies. Therefore, we are interested in estimating conditional market risk given observations of a large range of past financial data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Estimating market risk with neural networks

Franke¹,

Diagne²

2006

Statistics &Amp; Decisions

Self Cite

View full text Add to dashboard Cite

We consider stochastic volatility models for discrete financial time series of the nonlinear autoregressive-ARCH type with exogenous components. We discuss how the trend and volatility functions determining the process may be estimated nonparametrically by least-squares fitting of neural networks or, more generally, of functions from other parametric classes having a universal approximation property. We prove consistency of the estimates under conditions on the rate of increase of function complexity. The procedure is applied to the problem of quantifying market risk, i.e. of calculating volatility or value-at-risk from the data taking not only the time series of interest but additional market information into account. As an application, we study some stock prices series and compare our approach with the common method based on fitting a GARCH(1,1)-model to the data.

show abstract

Section: Neural Network Based Risk Managementmentioning

confidence: 94%

Section: Introductionmentioning

confidence: 99%

Estimating market risk with neural networks

Franke¹,

Diagne²

2006

Statistics &Amp; Decisions

Self Cite

View full text Add to dashboard Cite

show abstract

“…In particular, LSTM neural networks are state-of-the-art for many applications such as speech recognition, text extraction, translation or handwriting recognition, since plain recurrent neural networks are not suited to non-stationary time series modeling. In finance, Deep learning has been used in various research, in particular by Franke (1999) and Zhang et al (2020) for portfolio management or by (Kim and Won;2018) for volatility forecasting with LSTM in comparison with GARCH. Nevertheless, one has yet to prove the superiority of neural networks compared to simpler parametric models in their application.…”

Section: Neural Networkmentioning

confidence: 99%

Tail-risk protection: Machine Learning meets modern Econometrics

Spilak¹,

Härdle²

2020

Preprint

View full text Add to dashboard Cite

Tail risk protection is in the focus of the financial industry and requires solid mathematical and statistical tools, especially when a trading strategy is derived. Recent hype driven by machine learning (ML) mechanisms has raised the necessity to display and understand the functionality of ML tools. In this paper, we present a dynamic tail risk protection strategy that targets a maximum predefined level of risk measured by Value-At-Risk while controlling for participation in bull market regimes. We propose different weak classifiers, parametric and non-parametric, that estimate the exceedance probability of the risk level from which we derive trading signals in order to hedge tail events. We then compare the different approaches both with statistical and trading strategy performance, finally we propose an ensemble classifier that produces a meta tail risk protection strategy improving both generalization and trading performance.

show abstract

“…The nal inputs were selected using experience of expert traders and statistical model selection procedures. More details are given by F ranke 4 . The best network consisted of only H = 3 hidden neurons, but used 25-dimensional input vector x.…”

Section: Managing Portfolios Using Neural Networkmentioning

confidence: 99%