Using multi-stage data mining technique to build forecast model for Taiwan stocks

Huang, Chien-Jen; Chen, Peng-Wen; Pan, Wei

doi:10.1007/s00521-011-0628-0

Cited by 20 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For investors, the changes of stock index and stock price are very important information and the forecasting of time series in stock market has long been attracting the eternal interest of countless scientists and researchers for many years [9,17,[45][46][47]. Unfortunately, stock indices and prices are essentially dynamic, nonlinear, nonparametric and chaotic in nature.…”

Section: The Work Related To Time Series Forecastingmentioning

confidence: 99%

Forecasting stock index with multi-objective optimization model based on optimized neural network architecture avoiding overfitting

Zhou

Hou

Liu

2018

ComSIS

View full text Add to dashboard Cite

In this paper, the stock index time series forecasting using optimal neural networks with optimal architecture avoiding overfitting is studied. The problem of neural network architecture selection is a central problem in the application of neural network computation. After analyzing the reasons for overfitting and instability of neural networks, in order to find the optimal NNs (neural networks) architecture, we consider minimizing three objective indexes: training and testing root mean square error (RMSE) and testing error variance (TEV). Then we built a multi-objective optimization model, then converted it to single objective optimization model and proved the existence and uniqueness theorem of optimal solution. After determining the searching interval, a Multiobjective Optimization Algorithm for Optimized Neural Network Architecture Avoiding Overfitting (ONNAAO) is constructed to solve above model and forecast the time series. Some experiments with several different datasets are taken for training and forecasting. And some performance such as training time, testing RMSE and neurons, has been compared with the traditional algorithm (AR, ARMA, ordinary BP, SVM) through many numerical experiments, which fully verified the superiority, correctness and validity of the theory.

show abstract

Section: The Work Related To Time Series Forecastingmentioning

confidence: 99%

Forecasting stock index with multi-objective optimization model based on optimized neural network architecture avoiding overfitting

Zhou

Hou

Liu

2018

ComSIS

View full text Add to dashboard Cite

show abstract

“…It is adaptive, has good learning capability and has tolerated errors. Many scholars used this method to construct an early-warning model in financial distress, for example Altman et al [1], Coats and Fant [12], Huang [25], Nien [44], Odom and Sharda [45], Ohlson [46], Pan [47], Ravisankar and Ravi [50], Tai [57], Tam and Kiang [58], Theodossiou [59], Wu [62], and Yang et al [63].…”

Section: Case Studymentioning

confidence: 99%

“…All data were collected from the Taiwan Economic Journal Database from 1995 to 2009, a total of 285 samples. To improve the efficiency of the decisionmaking and analysis, first, we use the data mining-multivariate adaptive regression splines (MARS) to automatically assess the importance of each input variable for the forecasting task (Fayyad et al [17], Fayyad and Stolorz [18], Friedman [20], Huang et al [25], and Lewis and Stevens [33]). It extracts six key variables from these 96 input variables.…”

Section: Case Studymentioning

confidence: 99%

A novel 3D fruit fly optimization algorithm and its applications in economics

Lin

2015

Neural Comput & Applic

View full text Add to dashboard Cite

Fruit fly optimization algorithm (FOA) is a method that we have previously developed from the foodfinding behavior of fruit flies to solve optimization problems. The advantage of FOA is simple and easy to understand compared to traditional stochastic algorithms. In this paper, we propose a modified algorithm called novel 3D-FOA. The performances of the 3D-FOA are far better than those of the original FOA. We select more than thirty different nonlinear functions as test vehicles to show that the search efficiency and/or quality of the 3D-FOA is superior to that of the genetic algorithm and particle swarm optimization algorithm. We also apply the 3D-FOA on some economics topics, two theoretic examples and a case study. Our results strongly suggest that the 3D-FOA can enhance capabilities in a variety of fields and future experiments.

show abstract

“…Semantic access to databases has a long history and originated at early stages of database technology development. Unfortunately, they have not yet led to the creation of widely accepted industrial technologies [2]. Some of the researchers used a multi-stage optimized stock forecast model to grasp the changing trend of the stock market [7].…”

Section: Literature Reviewmentioning

confidence: 99%

Green Referential Data Base for the Indian Stock Market

Singh¹,

Dimri²,

Chakraborty³

2014

IJCA

View full text Add to dashboard Cite

A traditional database methodology has been used in the Indian stock market. Forecasting of the market is not only based on the prices of the stocks but also on other integrated information like socio-economic factors, prices, politics etc. Changing data format and its behavior requires a new methodology to handle and integrate. Investments solely depend upon efficiency and accuracy of the data. Data required for the stock market decision making process is generated from every event and all events have its own impact on the market. Irrespective of the nature of the event and its format, market requires data integration of all kind. Because of acute scarcity of natural resources, processing of the stock market data requires green methodologies which contribute to save energy, power, time, space etc. Fractal behavior of the market shows repetition of the stock prices again and again. Large amount of space, time, power etc have been utilized to store and process these repetitive data. Referential data base is one of the answers to this problem. This paper proposes referential data base for the stock market prices without compromising efficiency and accuracy for market forecasting methods.

show abstract

Using multi-stage data mining technique to build forecast model for Taiwan stocks

Cited by 20 publications

References 8 publications

Forecasting stock index with multi-objective optimization model based on optimized neural network architecture avoiding overfitting

Forecasting stock index with multi-objective optimization model based on optimized neural network architecture avoiding overfitting

A novel 3D fruit fly optimization algorithm and its applications in economics

Green Referential Data Base for the Indian Stock Market

Contact Info

Product

Resources

About