Financial Time Series Models Research Articles

Financial Sequence Prediction Based on Swarm Intelligence Algorithms of Internet of Things

In order to accurately predict the financial time series and stock price fluctuation, in this study, metadata was used to ensure the objectivity of data for most of the swarm intelligence algorithms of the Internet of Things. The results show that the combination of theoretical methods and financial time series forecasting model can significantly improve the forecasting effect. At the same time, the proposed fuzzy theory also gives a totally different perspective to the modeling of financial time series which has stochastic and uncertain characteristics. Therefore, swarm intelligence algorithm and fuzzy theory are effectively combined, and their combination is very useful for the improvement of two financial fuzzy time series models. Because of their combination, the problem of stock price forecasting with different orders and different factors is well solved. Besides, it also provides two new solutions for such problems respectively as well as additional model choices for investors to avoid risks and improve returns.

The Foreign Exchange Rate Prediction Using Long-Short Term Memory

The foreign exchange market is a global financial market that is influenced by economic, political, and psychological factors that are interconnected in complex ways. This complexity makes the foreign exchange market a difficult time-series prediction. At the end of 2019, the world was faced with the COVID-19 pandemic that has not only affected public health but also the foreign exchange market, which makes the trading behaviour affected. Long Short-Term Memory network (LSTM) is a type of recurrent neural network (RNN) that can solve long-term dependencies and is suitable to be a financial time-series model. This study implemented the LSTM model to predict the foreign exchange rate at a timeframe of 1 hour and daily in 2020 to get the best hyperparameter based on the RMSE evaluation results. Furthermore, with the obtained hyperparameters, the prediction result of 2020 was then compared with the 2018 and 2019 prediction results. The best RMSE result was obtained in 1-hour timeframe and when 2020’s RMSE result was compared to 2018’s and 2019’s RMSE result, the prediction of 2019 gave the best RMSE result. The LSTM model is able to achieve good results in the 2020 prediction, proven by the RMSE result which is 0.00135.

Financial time series prediction using wavelet and artificial neural network

This paper focuses on the modelling of financial time series using the coupling of the discrete wavelet transform and nonlinear autoregressive neural network. This hybrid modelling method is based on the use of decomposed time series using the discrete wavelet transform as inputs to the artificial neural networks. This method has been applied to three financial series (exchange rate EUR/USD, the Brent price and NASDAQ composite price). The simulation results using R software show the robustness of the proposed model compared to other modelling methods applied to the same financial time series.

A Quantum Mechanical Financial Time Series Model

This paper introduces a time series model derived from the reverse engineering of the Schrodinger equation. The model is a wave packet model which has a stochastic drift as a superposition of trigonometric functions. The model can be thought of as a financial time series model that includes a model of a real-world measure that allows the opportunity to profit from market timing strategies and a model of an efficient market hypothesis that does not allow it.

Wasserstein GAN: Deep Generation Applied on Financial Time Series

Modeling financial time series is challenging due to their high volatility and unexpected happenings on the market. Most financial models and algorithms trying to fill the lack of historical financial time series struggle to perform and are highly vulnerable to overfitting. As an alternative, we introduce in this paper a deep neural network called the WGAN-GP, a data-driven model that focuses on sample generation. The WGAN-GP consists of a generator and discriminator function which utilize an LSTM architecture. The WGAN-GP is supposed to learn the underlying structure of the input data, which in our case, is the Bitcoin. Bitcoin is unique in its behavior; the prices fluctuate what makes guessing the price trend hardly impossible. Through adversarial training, the WGAN-GP should learn the underlying structure of the bitcoin and generate very similar samples of the bitcoin distribution. The generated synthetic time series are visually indistinguishable from the real data. But the numerical results show that the generated data were close to the real data distribution but distinguishable. The model mainly shows a stable learning behavior. However, the model has space for optimization, which could be achieved by adjusting the hyperparameters.

Artificial chemical reaction optimisation of recurrent functional link neural networks for efficient modelling and forecasting of financial time series

In contrast to multilayer neural networks, a functional link artificial neural network uses functional expansion units for transferring lower input space to higher dimensions. It achieves enhanced ...

Artificial chemical reaction optimisation of recurrent functional link neural networks for efficient modelling and forecasting of financial time series

Artificial chemical reaction optimisation of recurrent functional link neural networks for efficient modelling and forecasting of financial time series

Testing for random coefficient autoregressive and stochastic unit root models

Abstract The random coefficient autoregressive model has been utilized for modeling financial time series because it possesses features that are often observed in financial time series. When the mean of the random coefficient is one, it is called the stochastic unit root model. This paper proposes two Lagrange multiplier tests for the null hypotheses of random coefficient autoregressive and stochastic unit root models against a more general model. We apply our Lagrange multiplier tests to several stock index data, and find that the stochastic unit root model is rejected, whereas the random coefficient autoregressive model is not. This result indicates that it is important to check the validity of the stochastic unit root model prior to applying it to financial time series data, which may be better modeled by the random coefficient autoregressive model with the mean being not equal to one.

Neural networks and their application in forecasting problems

The report describes popular machine learning methods and applications of neural networks. It reveals methods of training neural networks and offers a method of forecasting based on neural networks for modelling financial time series. Neural networks have recently gained in popularity among scholars. In particular, neural networks are widely used in the field of visualization and image recognition. But the practical significance of neural networks does not end there, they also find usage in such areas as forecasting, classification, clustering and modelling. This success can mainly be attributed to the remarkable property of models based on neural networks – they can «see» non-linear connections in contrast to many models, which for the most part have linear connections only. Currently, the use of neural networks is developing in the following directions: Stock market and macroeconomic forecasting (Neuro XL, OptimuStock, StocksNeural); Speech recognition and man-machine dialogue (Siri, Alexa, Cortana, Alice); Imitation of intellectual activity (weak AI in Siri, Alexa, Cortana, Alice); Improving low-quality and noisy information (DeepImagePrior). The advantages of using neural networks include: versatility. simplicity. Neural networks are able to model dependencies also in cases when there is a large number of variables.

Testing for Random Coefficient Autoregressive and Stochastic Unit Root Models

The random coefficient autoregressive model has been utilized for modeling financial time series because it possesses features that are often observed in financial time series. When the mean of the random autoregressive coefficient is one, it is called the stochastic unit root model. This paper proposes two Lagrange multiplier tests for the null hypothesis of random coefficient autoregressive and stochastic unit root models against a more general model. We apply our Lagrange multiplier tests to several stock index data, and find that the stochastic unit root model is rejected, whereas the random coefficient autoregressive model is not. This result indicates that it is important to check the validity of the stochastic unit root model prior to applying it to financial time series data, which may be better modeled by the random coefficient autoregressive model with the mean being not equal to one.

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

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.

Data mining model for multimedia financial time series using information entropy

At present, KDD research covers many aspects, and has achieved good results in the discovery of time series rules, association rules, classification rules and clustering rules. KDD has also been widely used in practical work such as OLAP and DW. Also, with the rapid development of network technology, KDD research based on WEB has been paid more and more attention. The main research content of this paper is to analyze and mine the time series data, obtain the inherent regularity, and use it in the application of financial time series transactions. In the financial field, there is a lot of data. Because of the huge amount of data, it is difficult for traditional processing methods to find the knowledge contained in it. New knowledge and new technology are urgently needed to solve this problem. The application of KDD technology in the financial field mainly focuses on customer relationship analysis and management, and the mining of transaction data is rare. The actual work requires a tool to analyze the transaction data and find its inherent regularity, to judge the nature and development trend of the transaction. Therefore, this paper studies the application of KDD in financial time series data mining, explores an appropriate pattern mining method, and designs an experimental system which includes mining trading patterns, analyzing the nature of transactions and predicting the development trend of transactions, to promote the application of KDD in the financial field.

ADAPTED BOX-COUNTING METHOD FOR ASSESSMENT OF THE FRACTAL DIMENSION OF FINANCIAL TIME SERIES

This paper continues research within the framework of the scientific direction in econophysics at the Department of Information Systems and Mathematical Methods in Economics of the faculty of Economics of PSU. Modeling and prediction of financial time series is quite a perspective area of research, because it allows participants of financial processes to reduce risks and make effective decisions. For example, we could research financial processes with the help of fractal analysis. In the article there is studied and worked out in detail one of the methods of fractal analysis of financial time series – the box-counting method for assessment of the fractal dimension. This method is often used in studies conducted by domestic authors, but the authors do not delve into the characteristics and problems of using the box-counting method for analysis of time series, that means that the answers to the interested questions have not yet been given. The main problem is that, as a rule, the analyzed object in the tasks of applying the box-counting method to time series is a computer image of the plot of series. In the article there is proposed the procedure of adaptation of the box-counting method for assessment of the fractal dimension of time series, the procedure does not require the formation of a computer image of the plot. In the article there is considered following difficulties developed from this adaptation: 1) high sensitivity of the resulting estimation of the dimension to the input parameters of the method (the ratio of the sides of the covered by cells plane with the plot; the used range of lengths of the side of the cell; the number of partitions of the plane into cells); 2) the non-obviousness of choosing the optimal values ​​of these parameters. In the article there are analyzed approaches to the selection of these parameters that were proposed by other authors, and there are determined the most suitable approaches for the adapted box-counting method. Also there are developed unique methods for determining the ratio of the sides of the plane with the plot. In the paper there is written the computer program that implements the developed method, and this program is tested on the generated data. The study obtained the following results. The fact of sensitivity of the adapted box-counting method to input parameters is confirmed, that indicates the high importance of the correct choice of these parameters. According to the study, there is found out inability of the proposed methods of automatic determination the ratio of the sides of the plane in relation to artificial time series. There are obtained the most precise (in a statistical sense) estimates of fractal dimension, those found by means of the adapted box-counting method, with the fixed ratio of the sides 1:1. According to comparing the adapted box-counting method and R/S analysis, there are obtained the most precise estimates by the second method (R/S analysis). Finally in the paper there are formulated the possible directions for further research: 1) comparison of the accuracy of various methods for assessment of the fractal dimension on series of different lengths; 2) comparison of the methods of fractal analysis and p-adic analysis for modeling and prediction of financial time series; 3) determination of the conditions of applicability of various methods; 4) approbation of the developed methods for determining of the ratio of the sides of the plane with the plot on real economic data.

Generating Financial Markets With Signatures

Neural network based data-driven market simulation unveils a new and flexible way of modelling financial time series, which has recently inspired a surge of research activity in the quantitative finance community. Though generative market simulation is model-free in the sense that it makes no assumptions on the stochastic dynamics of the underlying paths, the concrete modelling choices are nevertheless decisive for the performance of the resulting market generators and the features of the simulated paths. We contrast some classical approaches of market simulation with simulation based on generative modelling and highlight some advantages and pitfalls of the new approach. While most generative models tend to rely on large amounts of training data, we present here a generative model that works reliably even in environments where the amount of available training data is small, irregularly paced or oscillatory. We show how a rough paths-based feature map encoded by the signature of the path outperforms returns based market generation both numerically and from a theoretical point of view. Finally, we also propose a suitable performance evaluation metric for financial time series and discuss some connections of our signature-based Market Generator to deep hedging.

Cepstral-based clustering of financial time series

In this paper, following the Partitioning Around Medoids (PAM) approach and the fuzzy theory, we propose a clustering model for financial time series based on the estimated cepstrum which represents the spectrum of the logarithm of the spectral density function. Selecting the optimal set of financial securities to build a portfolio that aims to maximize the risk-return tradeoff is a largely investigated topic in finance. The proposed model inherits all the advantages connected to PAM approach and fuzzy theory and it is able to compute objectively the cepstral weight associated to each cepstral coefficient by means of a suitable weighting system incorporated in the clustering model. In this way, the clustering model is able to tune objectively the different influence of each cepstral coefficient in the clustering process. The proposed clustering model performs better with respect to other clustering models. The proposed clustering model applied to each security sharpe ratio provides an efficient tool of clustering of stocks.

On the ARCH model with stationary liquidity

The classical ARCH model together with its extensions have been widely applied in the modeling of financial time series. We study a variant of the ARCH model that takes account of liquidity given by a positive stationary process. We provide minimal assumptions that ensure the existence and uniqueness of the stationary solution for this model. Moreover, we give necessary and sufficient conditions for the existence of the autocovariance function. After that, we derive an AR(1) characterization for the stationary solution yielding Yule–Walker type quadratic equations for the model parameters. In order to define a proper estimation method for the model, we first show that the autocovariance estimators of the stationary solution are consistent under relatively mild assumptions. Consequently, we prove that the natural estimators arising out of the quadratic equations inherit consistency from the autocovariance estimators. Finally, we illustrate our results with several examples and a simulation study.

Quant GANs: deep generation of financial time series

Modeling financial time series by stochastic processes is a challenging task and a central area of research in financial mathematics. As an alternative, we introduce Quant GANs, a data-driven model which is inspired by the recent success of generative adversarial networks (GANs). Quant GANs consist of a generator and discriminator function, which utilize temporal convolutional networks (TCNs) and thereby achieve to capture long-range dependencies such as the presence of volatility clusters. The generator function is explicitly constructed such that the induced stochastic process allows a transition to its risk-neutral distribution. Our numerical results highlight that distributional properties for small and large lags are in an excellent agreement and dependence properties such as volatility clusters, leverage effects, and serial autocorrelations can be generated by the generator function of Quant GANs, demonstrably in high fidelity.

Application of grey relational analysis and artificial neural networks on currency exchange-traded notes (ETNs)

Abstract This study determines which index has the strongest influence on the exchange-trade note (ETN) returns using the grey relational analysis. Results show that the volatility index is the strongest, followed by the S&P 500 stock index, the US dollar index, the CRB index, the Trade index, and the Brent crude oil index. However, the US dollar index has the most significant effect of using the index values of currency ETNs, followed by the S&P 500 stock index, volatility index, Brent crude oil index, the CRB index, and Trade index. This study applies four types of the artificial neural network model, namely, back-propagation neural network (BPN), recurrent neural network (RNN), time-delay recurrent neural network (TDRNN), and radial basis function neural network (RBFNN) to capture the nonlinear tendencies of ETNs for better forecasting accuracy. The paper finds that the RNN and RBFNN models have stronger predictive power among the models, and provides the highest forecasting accuracy for the majority of the currency ETNs. However, the RNN model consistently shows that the low grey relational grades (GRG) variables have the strongest influence on the ETN returns, compared with combining all and high GRG variables. These findings suggest that fund managers and traders can potentially rely on both RNN and RBFNN models, particularly the former, in their applications in financial time-series modeling.

Forecast model for financial time series: An approach based on harmonic oscillators

A financial asset price-forecasting model based on damped driven harmonic oscillator is presented. It is inspired on the idea that in the price fluctuations a restoring force to a supposed fair price, besides inertia and dissipation, could take place. The model is evaluated in an emerging market, which is of the Brazilian stock exchange BM&FBovespa (B3). The choice is due to the expectation that the model could perform better by exploiting deviations from the concept of efficient market. In such direction, the use of Hurst exponent to choose suitable assets to trade employing the model is discussed. A fairly common trading system endowed with the stop gain and stop loss limiting parameters was used to evaluate the model. To avoid excessive arbitrariness, those parameters were established after a study referred to the data itself in which a consistently defined relation return/risk is computed. The proposed model overcomes a full random model in four studied cases. In the aggregated result of 5 stocks, it was achieved a return of investment of 64:80%, disregarding operating costs. This is 72.19% higher than the respective result from random model.

A Hybrid Model for Financial Time Series Forecasting—Integration of EWT, ARIMA With The Improved ABC Optimized ELM

The practical significances and complexities of financial time series analysis induce highly demand more reliable hybrid model that denoised the data efficiently, handled with both linear and nonlinear patterns in the data, to achieve more accurate results. This paper suggests a new forecasting hybrid model for financial time series data combined Empirical Wavelet Transform (EWT) technique with improved Artificial Bee Colony (ABC) algorithm, Extreme Learning Machine (ELM) neural network, and Auto-Regressive Integrated Moving Average (ARIMA) linear analysis algorithm. The EWT is used to decompose and denoise the data to reconstruct the data more suitable for forecast. The improvement of the ABC algorithm is according to the Good Point Sets (GPS) theory and adaptive Elite-based Opposition (EO) strategy (GPS-EO-ABC) to overcome the drawbacks of the original algorithm and enhance the optimization performance. The optimized ELM with GPS-EO-ABC, as well as the ARIMA, are utilized independently to generated different forecasting results and combined by the weight-based procession. We testify the performance of the proposed improved ABC algorithm by ten benchmark functions, simulating the proposed forecasting models by three financial time-series datasets. The results indicate that: (1) The proposed algorithm shows outstanding capacities in parameter optimization. The optimized ELM generated more stable and precise results compared with original ELM, ABC-ELM, single LSTM, and ANN; (2) The proposed hybrid model has not only effectiveness but also efficiencies in denoising data, correcting outliers, coordinating both linear and nonlinear patterns, its performance in financial time series forecasting is more excellent than existing hybrid models.