An Effective Stock Classification Method Via MDS Based on Modified Mutual Information Distance

2019 ◽  
Vol 19 (02) ◽  
pp. 2050018 ◽  
Author(s):  
Jun Jiang ◽  
Pengjian Shang ◽  
Xuemei Li
Keyword(s):  
Time Series ◽  
Mutual Information ◽  
Stock Market ◽  
Financial Data ◽  
Simulation Experiments ◽  
Market Data ◽  
Information Distance ◽  
Proximity Matrix ◽  
The Stability ◽  
Stability And Accuracy

This paper proposes a multidimensional scaling (MDS) method based on modified mutual information distance (M-MDS) to analyze stock market data. To better describe the relativity of financial data, it is worthwhile to point out that the commonly used proximity matrix in MDS is replaced with modified mutual information distance (M-MI-D) matrix. Refer to M-MI-D, a higher dissimilarity leads to a larger distance. In order to demonstrate the stability and accuracy of M-MDS, logistic time series are used in simulation experiments. In addition, a comparison of this new M-MDS method with classical MDS is given using the stock market data. It is noted that the new M-MDS method shows better stability than that of classical MDS method. Moreover, not only the stocks in the same US stock block, but also the stocks in different blocks have been discussed to illustrate the efficiency of M-MDS method.

Pemodelan Data Time Series Garch(1,1) Untuk Pasar Saham Indonesia

2015 ◽  
Vol 11 (1) ◽  
pp. 13
Author(s):  
Elfa Rafulta ◽  
Roni Tri Putra
Keyword(s):  
Time Series ◽  
Stock Market ◽  
Stock Markets ◽  
Financial Data ◽  
International Stock Markets

This paper introduced a method pengklusteran for financial data. By using the model Heteroskidastity Generalized autoregressive conditional (GARCH), will be estimated distance between the stock market using GARCH-based distance. The purpose of this method is mengkluster international stock markets with different amounts of data.

scholarly journals Deep Learning-based Integrated Stacked Model for the Stock Market Prediction

2019 ◽  
Vol 9 (1) ◽  
pp. 5167-5174 ◽  
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Deep Learning ◽  
Stock Market ◽  
Network Models ◽  
Time Series Forecasting ◽  
Convolution Neural Network ◽  
Neural Network Models ◽  
Stock Market Prediction ◽  
Market Data

Recently, the stock market prediction has become one of the essential application areas of time-series forecasting research. The successful prediction of the stock market can be better guided to the investors to maximize their profit and to minimize the risk of investment. The stock market data are very much complex, non-linear and dynamic. Due to this reason, still, it is a challenging task. In recent time, deep learning method has become one of the most popular machine learning methods for time-series forecasting due to their temporal feature extraction capabilities. In this paper, we have proposed a novel Deep Learning-based Integrated Stacked Model (DISM) that integrates both the 1D Convolution neural network and LSTM recurrent neural network to find the spatial and temporal features from the stock market data. Our proposed DISM is applied to forecast the stock market. Here, we have also compared our proposed DISM with the single structured stacked LSTM, and 1D Convolution neural network models, and some other statistical models. We have observed that our proposed DISM produces better results in terms of accuracy and stability.

Interaction Between Forex and Stock Markets in India: An Entropy Approach

2008 ◽  
Vol 33 (4) ◽  
pp. 27-46 ◽  
Author(s):  
Y V Reddy ◽  
A Sebastin
Keyword(s):  
Time Series ◽  
Mutual Information ◽  
Stock Market ◽  
Transfer Entropy ◽  
The Other ◽  
Stock Index ◽  
Reverse Direction ◽  
Probability Mass Function ◽  
Information Theoretic ◽  
The Relationship

Interactions between the foreign exchange market and the stock market of a country are considered to be an important internal force of the markets in a financially liberalized environment. If causal relationship from a market to the other is not detected, then informational efficiency exists in the other whereas existence of causality implies that hedging of exposure to one market by taking position in the other market will be effective. The temporal relationship between the forex market and the stock market of developing and developed countries has been studied, especially after the East Asian financial crisis of 1997–98, using various methods like cross-correlation, cross-spectrum, and error correction model, but these methods identify only linear relations. A statistically rigorous approach to the detection of interdependence, including non-linear dynamic relationships, between time series is provided by tools defined using the information theoretic concept of entropy. Entropy is the amount of disorder in the system and also is the amount of information needed to predict the next measurement with a certain precision. The mutual information between two random variables X and Y with a joint probability mass function p(x,y) and marginal mass functions p(x) and p(y), is defined as the relative entropy between the joint distribution p(x,y) and the product distribution p(x)*p(y). Mutual information is the reduction in the uncertainty of X due to the knowledge of Y and vice versa. Since mutual information measures the deviation from independence of the variables, it has been proposed as a tool to measure the relationship between financial market segments. However, mutual information is a symmetric measure and does not contain either dynamic information or directional sense. Even time delayed mutual information does not distinguish information actually exchanged from shared information due to a common input signal or history and therefore does not quantify the actual overlap of the information content of two variables. Another information theoretic measure called transfer entropy has been introduced by Thomas Schreiber (2000) to study the relationship between dynamic systems; the concept has also been applied by some authors to study the causal structure between financial time series. In this paper, an attempt has been made to study the interaction between the stock and the forex markets in India by computing transfer entropy between daily data series of the 50 stock index of the National Stock Exchange of India Limited, viz., Nifty and the exchange rate of Indian Rupee vis- à- vis US Dollar, viz., Reserve Bank of India reference rate. The entire period–November 1995 to March 2007–selected for the study, has been divided into three sub-periods for the purpose of analysis, considering the developments that took place during these sub-periods. The results obtained reveal that: there exist only low level interactions between the stock and the forex markets of India at a time scale of a day or less, although theory suggests interactive relationship between the two markets the flow from the stock market to the forex market is more pronounced than the flow in the reverse direction.

scholarly journals Fuzzy Transfer Learning in Time Series Forecasting for Stock Market Prices

2021 ◽  
Author(s):  
Shanoli Samui Pal ◽  
Samarjit Kar
Keyword(s):  
Time Series ◽  
Stock Market ◽  
Prior Knowledge ◽  
Transfer Learning ◽  
Prediction Accuracy ◽  
Time Series Data ◽  
Time Series Prediction ◽  
Series Data ◽  
Data Sets ◽  
Market Data

Abstract Transfer learning involves transferring prior knowledge of solving similar problems in order to achieve quick and efficient solution. The aim of fuzzy transfer learning is to transfer prior knowledge in an imprecise environment. Time series like stock market data are non-linear in nature and movement of stock is uncertain, so it is quite difficult following the stock market and in decision making. In this study, we propose a method to forecast stock market time series in the situation when we can use prior experience to make decisions. Fuzzy transfer learning (FuzzyTL) is based on knowledge transfer in that and adapting rules obtained domain. Three different stock market time series data sets are used for comparative study. It is observed that the effect of knowledge transferring works well together with smoothing of dependent attributes as the stock market data fluctuate with time. Finally, we give an empirical application in Shenzhen stock market with larger data sets to demonstrate the performance of the model. We have explored FuzzyTL in time series prediction to unerstand the essence of FuzzyTL. We were working on the question of the capability of FuzzyTL in improving prediction accuracy. From the comparisons, it can be said fuzzy transfer learning with smoothing improves prediction accuracy efficiently.

Stock market data modelling using fitness-oriented JAYA algorithm-based deep belief network

Kybernetes ◽  
2019 ◽  
Vol 49 (9) ◽  
pp. 2309-2334
Author(s):  
A. Kullaya Swamy ◽  
Sarojamma B.
Keyword(s):  
Time Series ◽  
Stock Market ◽  
Fitness Function ◽  
Error Function ◽  
Percentage Error ◽  
Jaya Algorithm ◽  
Market Data ◽  
Content Type ◽  
Proposed Model ◽  
L2 Norm

Purpose Data mining plays a major role in forecasting the open price details of the stock market. However, it fails to address the dimensionality and expectancy of a naive investor. Hence, this paper aims to study a future prediction model named time series model is implemented. Design/methodology/approach In this model, the stock market data are fed to the proposed deep neural networks (DBN), and the number of hidden neurons is optimized by the modified JAYA Algorithm (JA), based on the fitness function. Hence, the algorithm is termed as fitness-oriented JA (FJA), and the proposed model is termed as FJA-DBN. The primary objective of this open price forecasting model is the minimization of the error function between the modeled and actual output. Findings The performance analysis demonstrates that the deviation of FJA–DBN in predicting the open price details of the Tata Motors, Reliance Power and Infosys data shows better performance in terms of mean error percentage, symmetric mean absolute percentage error, mean absolute scaled error, mean absolute error, root mean square error, L1-norm, L2-Norm and Infinity-Norm (least infinity error). Research limitations/implications The proposed model can be used to forecast the open price details. Practical implications The investors are constantly reviewing past pricing history and using it to influence their future investment decisions. There are some basic assumptions used in this analysis, first being that everything significant about a company is already priced into the stock, other being that the price moves in trends Originality/value This paper presents a technique for time series modeling using JA. This is the first work that uses FJA-based optimization for stock market open price prediction.

scholarly journals SEPARACJA FINANSOWYCH SZEREGÓW CZASOWYCH Z WYKORZYSTANIEM DEKORELACJI Z OPÓŹNIENIAMI

2020 ◽  
Vol 20 (4) ◽  
pp. 305-315
Author(s):  
Ryszard Szupiluk ◽  
Paweł Rubach
Keyword(s):  
Time Series ◽  
Stock Market ◽  
Financial Time Series ◽  
Methodological Issues ◽  
Market Data ◽  
Financial Time ◽  
Stock Indices

In this paper, we present the separation of financial time series using algorithms based on the decoralation procedure. The SOBI and AMUSE algorithms are used, tested and compared on real stock market data. We also present a discussion of theoretical and methodological issues related to the application of separation algorithms. The study is carried out using the WIG20 and SP500 stock indices.

scholarly journals Development of Stock Networks Using Part Mutual Information and Australian Stock Market Data

Entropy ◽  
2020 ◽  
Vol 22 (7) ◽  
pp. 773
Author(s):  
Yan Yan ◽  
Boyao Wu ◽  
Tianhai Tian ◽  
Hu Zhang
Keyword(s):  
Information Theory ◽  
Big Data ◽  
Mutual Information ◽  
Stock Market ◽  
Correlation Coefficient ◽  
Market Data ◽  
Industrial Sectors ◽  
Large Size ◽  
Networks Analysis ◽  
The Relationship

Complex network is a powerful tool to discover important information from various types of big data. Although substantial studies have been conducted for the development of stock relation networks, correlation coefficient is dominantly used to measure the relationship between stock pairs. Information theory is much less discussed for this important topic, though mutual information is able to measure nonlinear pairwise relationship. In this work we propose to use part mutual information for developing stock networks. The path-consistency algorithm is used to filter out redundant relationships. Using the Australian stock market data, we develop four stock relation networks using different orders of part mutual information. Compared with the widely used planar maximally filtered graph (PMFG), we can generate networks with cliques of large size. In addition, the large cliques show consistency with the structure of industrial sectors. We also analyze the connectivity and degree distributions of the generated networks. Analysis results suggest that the proposed method is an effective approach to develop stock relation networks using information theory.

scholarly journals Simple Model for Market Returns Distribution

2013 ◽  
Vol 23 (2) ◽  
pp. 185 ◽  
Author(s):  
Chu Thuy Anh ◽  
Do Hong Lien ◽  
Nguyen Ai Viet
Keyword(s):  
Relaxation Time ◽  
Simple Model ◽  
Stock Market ◽  
Large Time ◽  
Boltzmann Distribution ◽  
Time Parameter ◽  
Market Returns ◽  
Stock Market Returns ◽  
Market Data ◽  
The Stability

It has been observed that at the large time scales the distributionof stock market returns is convergent from Boltzmann distribution to Gaussianasymptotic one. To explain this universal phenomenon, we propose a new andsimple dynamic model to describe this convergence by the time parameter inassociation with the introducing the concept of relaxation time for nancialmarkets. The analysis of stock market data packages in dierent time intervalsshowed that our model ts well the nancial market data. The meaning ofso{called relaxation time has been qualitatively made clear, as a measure toestimate the stability of the market.

Sign in / Sign up

Export Citation Format

Share Document