Increasing Translation Invariant Morphological Forecasting Models for Stock Market Prediction

2010 ◽  
pp. 174-210
Author(s):  
Ricardo de A. Araújo
Keyword(s):  
Time Series ◽  
Performance Metrics ◽  
Financial Time Series ◽  
Back Propagation ◽  
Original Data ◽  
Translation Invariant ◽  
Financial Time ◽  
Proposed Model ◽  
Classical Models ◽  
Takens Theorem

Statistical models have been widely used for the purpose of forecasting. However, it has some limitations regarding its performance, which prevents an automatic forecasting system development. In order to overcome such limitations, Artificial Neural Networks (ANNs), Evolutionary Algorithms (EAs) and Fuzzy Systems (FSs) based approaches have been proposed for nonlinear time series modeling. However, a dilemma arises from all these models regarding financial time series, which follow a Random Walk (RW) model, where the forecast of such time series exhibits a characteristic one step shift regarding original data. In this way, this work presents a new approach, referred to as Increasing Translation Invariant Morphological Forecasting (ITIMF) model, to overcome the RW dilemma for financial time series forecasting, which performs an evolutionary search for the minimum dimension to determining the characteristic phase space that generates the financial time series phenomenon. It is inspired on Takens Theorem and consists of an intelligent hybrid model composed of a Modular Morphological Neural Network (MMNN) combined with a Modified Genetic Algorithm (MGA), which searches for the particular time lags capable of a fine tuned characterization of the time series and estimates the initial (sub-optimal) parameters (weights, architecture and number of modules) of the MMNN. Each individual of the MGA population is trained by the Back Propagation (BP) algorithm to further improve the MMNN parameters supplied by the MGA. After adjusting the model, it performs a behavioral statistical test and a phase fix procedure to adjust time phase distortions observed in financial time series. Furthermore, an experimental analysis is conducted with the proposed model using ten real world financial time series. Five well-known performance metrics and an evaluation function are used to assess the performance of the proposed model and the obtained results are compared to classical models presented in literature.

scholarly journals A Financial Embedded Vector Model and Its Applications to Time Series Forecasting

2018 ◽  
Vol 13 (5) ◽  
pp. 881-894
Author(s):  
Yanfeng Sun ◽  
Minglei Zhang ◽  
Si Chen ◽  
Xiaohu Shi
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Language Processing ◽  
Prediction Accuracy ◽  
Financial Time Series ◽  
Vector Model ◽  
Rbf Neural Networks ◽  
Vector Representation ◽  
Financial Time ◽  
Proposed Model

Inspired by the embedding representation in Natural Language Processing (NLP), we develop a financial embedded vector representation model to abstract the temporal characteristics of financial time series. Original financial features are discretized firstly, and then each set of discretized features is considered as a “word” of NLP, while the whole financial time series corresponds to the “sentence” or “paragraph”. Therefore the embedded vector models in NLP could be applied to the financial time series. To test the proposed model, we use RBF neural networks as regression model to predict financial series by comparing the financial embedding vectors as input with the original features. Numerical results show that the prediction accuracy of the test data is improved for about 4-6 orders of magnitude, meaning that the financial embedded vector has a strong generalization ability.

scholarly journals GO-GJRSK Model with Application to Higher Order Risk-Based Portfolio

Mathematics ◽  
2020 ◽  
Vol 8 (11) ◽  
pp. 1990
Author(s):  
Kei Nakagawa ◽  
Yusuke Uchiyama
Keyword(s):  
Time Series ◽  
Stock Prices ◽  
Financial Time Series ◽  
Higher Order ◽  
Estimation Methods ◽  
Time Series Modeling ◽  
Higher Order Moments ◽  
Financial Time ◽  
Proposed Model ◽  
Multivariate Financial Time Series

There are three distinguishing features in the financial time series, such as stock prices, are as follows: (1) Non-normality, (2) serial correlation, and (3) leverage effect. All three points need to be taken into account to model the financial time series. However, multivariate financial time series modeling involves a large number of stocks, with many parameters to be estimated. Therefore, there are few examples of multivariate financial time series modeling that explicitly deal with higher-order moments. Furthermore, there is no multivariate financial time series model that takes all three characteristics above into account. In this study, we propose the generalized orthogonal (GO)-Glosten, Jagannathan, and Runkle GARCH (GJR) model which extends the GO-generalized autoregressive conditional heteroscedasticity (GARCH) model and incorporates the three features of the financial time series. We confirm the effectiveness of the proposed model by comparing the performance of risk-based portfolios with higher-order moments. The results show that the performance with our proposed model is superior to that with baseline methods, and indicate that estimation methods are important in risk-based portfolios with higher moments.

scholarly journals A Trend-Switching Financial Time Series Model with Level-Duration Dependence

2012 ◽  
Vol 2012 ◽  
pp. 1-20
Author(s):  
Qingsheng Wang ◽  
Aifan Ling ◽  
Tao Huang ◽  
Yong Jiang ◽  
Min Chen
Keyword(s):  
Time Series ◽  
Point Process ◽  
Turning Point ◽  
Financial Time Series ◽  
Difficult Problem ◽  
Time Series Model ◽  
Financial Time ◽  
Proposed Model ◽  
Long Time ◽  
Local Trend

The financial time series model that can capture the nonlinearity and asymmetry of stochastic process has been paid close attention for a long time. However, it is still open to completely overcome the difficult problem that motivates our researches in this paper. An asymmetric and nonlinear model with the change of local trend depending on local high-low turning point process is first proposed in this paper. As the point process can be decomposed into the two different processes, a high-low level process and an up-down duration process, we then establish the so-called trend-switching model which depends on both level and duration (Trend-LD). The proposed model can predict efficiently the direction and magnitude of the local trend of a time series by incorporating the local high-low turning point information. The numerical results on six indices in world stock markets show that the proposed Trend-LD model is suitable for fitting the market data and able to outperform the traditional random walk model.

An Evolutionary Functional Link Neural Fuzzy Model for Financial Time Series Forecasting

2013 ◽  
pp. 189-205
Author(s):  
S. Chakravarty ◽  
P. K. Dash ◽  
V. Ravikumar Pandi ◽  
B. K. Panigrahi
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Learning Algorithm ◽  
Financial Time Series ◽  
Back Propagation ◽  
Fuzzy Model ◽  
Series Data ◽  
Functional Link ◽  
Financial Time ◽  
Neural Fuzzy

This paper proposes a hybrid model, evolutionary functional link neural fuzzy model (EFLNF), to forecast financial time series where the parameters are optimized by two most efficient evolutionary algorithms: (a) genetic algorithm (GA) and (b) particle swarm optimization (PSO). When the periodicity is just one day, PSO produces a better result than that of GA. But the gap in the performance between them increases as periodicity increases. The convergence speed is also better in case of PSO for one week and one month a head prediction. To testify the superiority of the EFLNF, a number of comparative studies have been made. First, functional link artificial neural network (FLANN) and functional link neural fuzzy (FLNF) were combined with back propagation (BP) learning algorithm. The result shows that FLNF performs better than FLANN. Again, FLNF is compared with EFLNF where the latter outperforms the former irrespective of the periodicity or the learning algorithms with which it has been combined. All models are used to predict the most chaotic financial time series data; BSE Sensex and S&P CNX Nifty stock indices one day, one week and one month in advance.

A spectral analysis based heteroscedastic model for the estimation of value at risk

2018 ◽  
Vol 19 (3) ◽  
pp. 295-314
Author(s):  
Yang Zhao
Keyword(s):  
At Risk ◽  
Time Series ◽  
Spectral Analysis ◽  
Value At Risk ◽  
Financial Time Series ◽  
Estimation Method ◽  
Reliable Estimate ◽  
Content Type ◽  
Financial Time ◽  
Proposed Model

Purpose This paper aims to focus on a better model to capture the trait of varying volatility in various financial time series, as well as to obtain reliable estimate of value at risk (VaR). Design/methodology/approach The typical methods in spectral analysis are used to obtain the sample of conditional mean, conditional variance and residual term. The generalized regression neural network is used to establish a time-varying non-linear model, and the non-parametric kernel density estimation method is applied for the estimation of VaR. Findings The proposed model is able to follow the heteroscedastic characteristic which is common in financial time series, and the estimated VaR is satisfactory. Practical implications The analysis method in this study allows the hedgers, bankers, financial analysts as well as economists to draw a better inference from financial time series. Also, relatively more precise estimate of the VaR value for a certain kind of financial asset is available. The model with its derived estimates would definitely help in developing other models. Originality/value Up-to-date, the study in literature which models financial time serial from the viewpoint of spectral analysis is rare to see. Thus, the proposed model, along with a comprehensive empirical study which reveals desirable result for the estimation of VaR would enrich the related researches at present.

An Evolutionary Functional Link Neural Fuzzy Model for Financial Time Series Forecasting

2011 ◽  
Vol 2 (3) ◽  
pp. 39-58 ◽  
Author(s):  
S. Chakravarty ◽  
P. K. Dash ◽  
V. Ravikumar Pandi ◽  
B. K. Panigrahi
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Learning Algorithm ◽  
Financial Time Series ◽  
Back Propagation ◽  
Fuzzy Model ◽  
Series Data ◽  
Functional Link ◽  
Financial Time ◽  
Neural Fuzzy

This paper proposes a hybrid model, evolutionary functional link neural fuzzy model (EFLNF), to forecast financial time series where the parameters are optimized by two most efficient evolutionary algorithms: (a) genetic algorithm (GA) and (b) particle swarm optimization (PSO). When the periodicity is just one day, PSO produces a better result than that of GA. But the gap in the performance between them increases as periodicity increases. The convergence speed is also better in case of PSO for one week and one month a head prediction. To testify the superiority of the EFLNF, a number of comparative studies have been made. First, functional link artificial neural network (FLANN) and functional link neural fuzzy (FLNF) were combined with back propagation (BP) learning algorithm. The result shows that FLNF performs better than FLANN. Again, FLNF is compared with EFLNF where the latter outperforms the former irrespective of the periodicity or the learning algorithms with which it has been combined. All models are used to predict the most chaotic financial time series data; BSE Sensex and S&P CNX Nifty stock indices one day, one week and one month in advance.

Sign in / Sign up

Export Citation Format

Share Document