scholarly journals A Novel Approach to Short-Term Stock Price Movement Prediction using Transfer Learning

2019 ◽  
Vol 9 (22) ◽  
pp. 4745 ◽  
Author(s):  
Thi-Thu Nguyen ◽  
Seokhoon Yoon
Keyword(s):  
Time Series ◽  
Stock Market ◽  
Transfer Learning ◽  
Stock Price ◽  
The United States ◽  
Short Term ◽  
Price Movement ◽  
Stock Price Prediction ◽  
Training Samples ◽  
Korean Market

Stock price prediction has always been an important application in time series predictions. Recently, deep neural networks have been employed extensively for financial time series tasks. The network typically requires a large amount of training samples to achieve high accuracy. However, in the stock market, the number of data points collected on a daily basis is limited in one year, which leads to insufficient training samples and accordingly results in an overfitting problem. Moreover, predicting stock price movement is affected by various factors in the stock market. Therefore, choosing appropriate input features for prediction models should be taken into account. To address these problems, this paper proposes a novel framework, named deep transfer with related stock information (DTRSI), which takes advantage of a deep neural network and transfer learning. First, a base model using long short-term memory (LSTM) cells is pre-trained based on a large amount of data, which are obtained from a number of different stocks, to optimize initial training parameters. Second, the base model is fine-tuned by using a small amount data from a target stock and different types of input features (constructed based on the relationship between stocks) in order to enhance performance. Experiments are conducted with data from top-five companies in the Korean market and the United States (US) market from 2012 to 2018 in terms of the highest market capitalization. Experimental results demonstrate the effectiveness of transfer learning and using stock relationship information in helping to improve model performance, and the proposed approach shows remarkable performance (compared to other baselines) in terms of prediction accuracy.

scholarly journals Systematic optimization of Long Short-Term Memory model for predicting NYSE Arca Airline Index (XAL) during COVID-19

2021 ◽  
Vol 10 (1) ◽  
Author(s):  
Sarah Dong ◽  
Amber Wang
Keyword(s):  
Stock Market ◽  
Stock Prices ◽  
Airline Industry ◽  
Stock Price ◽  
Short Term Memory ◽  
The United States ◽  
Short Term ◽  
Term Memory ◽  
Stock Price Prediction ◽  
Long Short Term Memory

Predicting stock prices has been both challenging and controversial. Since it first spread through the United States, the COVID-19 pandemic has impacted the stock market in a multitude of ways. Thus, stock price prediction has become even more challenging. Recurrent neural networks (RNN) have been widely used in many fields to predict financial time series. In this study, Long Short-Term Memory (LSTM), a special form of RNN, is used to predict the stock market direction for the US airline industry by using NYSE Arca Airline Index (XAL). The LSTM model was optimized through changing different hyperparameters of the model architecture to find the best combination for increased accuracy and performance evaluated by several metrics, including raw RMSE (3.51) and MAPA (4.6%), and very high MAPA (95.4%) and R^2 (0.978).

scholarly journals An Economic Forecasting Method Based on the LightGBM-Optimized LSTM and Time-Series Model

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Jiehua Lv ◽  
Chao Wang ◽  
Wei Gao ◽  
Qiumin Zhao
Keyword(s):  
Decision Making ◽  
Time Series ◽  
Stock Market ◽  
Stock Prices ◽  
Stock Price ◽  
Large Scale ◽  
Stock Index ◽  
Short Term ◽  
Economic Forecasting ◽  
Stock Price Prediction

Stock price prediction is very important in financial decision-making, and it is also the most difficult part of economic forecasting. The factors affecting stock prices are complex and changeable, and stock price fluctuations have a certain degree of randomness. If we can accurately predict stock prices, regulatory authorities can conduct reasonable supervision of the stock market and provide investors with valuable investment decision-making information. As we know, the LSTM (Long Short-Term Memory) algorithm is mainly used in large-scale data mining competitions, but it has not yet been used to predict the stock market. Therefore, this article uses this algorithm to predict the closing price of stocks. As an emerging research field, LSTM is superior to traditional time-series models and machine learning models and is suitable for stock market analysis and forecasting. However, the general LSTM model has some shortcomings, so this paper designs a LightGBM-optimized LSTM to realize short-term stock price forecasting. In order to verify its effectiveness compared with other deep network models such as RNN (Recurrent Neural Network) and GRU (Gated Recurrent Unit), the LightGBM-LSTM, RNN, and GRU are respectively used to predict the Shanghai and Shenzhen 300 indexes. Experimental results show that the LightGBM-LSTM has the highest prediction accuracy and the best ability to track stock index price trends, and its effect is better than the GRU and RNN algorithms.

scholarly journals STOCK PRICE PREDICTION USING MARKOV CHAINS ANALYSIS WITH VARYING STATE SPACE ON DATA FROM THE CZECH REPUBLIC

2021 ◽  
Vol 24 (4) ◽  
pp. 142-155
Author(s):  
Milan Svoboda ◽  
Pavla Říhová
Keyword(s):  
Standard Deviation ◽  
Stock Market ◽  
Markov Chains ◽  
State Space ◽  
Stock Price ◽  
Business Models ◽  
Short Term ◽  
Stock Price Prediction ◽  
Price Prediction ◽  
Transaction Fees

The article describes empirical research that deals with short-term stock price prediction. The aim of this study is to use this prediction to create successful business models. A business model that outperforms the stock market, represented by the Buy and Hold strategy, is considered to be successful. A stochastic model based on Markov chains analysis with varying state space is used for short-term stock price prediction. The varying state spate is defined based on multiples of the moving standard deviation. A total of 80 state space models were calculated for the moving standard deviation with 5-step lengths from 10 to 30 in combination with the standard deviation multiples from 0.5 to 2.0 with the step of 0.1. The efficiency of the business models was verified for 3 long-term, liquid stocks of the Czech stock market, namely the stocks of KB, CEZ, and O2 within a 14-year period – from the beginning of 2006 to the end of 2019. Business models perform best when they use a state space defined on the length of a moving standard deviation between 15 and 30 in combination with multiples of the standard deviation between 1.1 and 1.2. Business models based on these parameters outperform the passive Buy and Hold strategy. In fact, they outperform the Buy and Hold strategy for both the entire period under review and the yielded five-year periods (including transaction fees). The only exception is the five-year periods covering 2015 for O2 stocks. After the end of the uncertainty period caused by unclear intentions of the new majority stockholder, the stock price rose sharply. These results are in conflict with the efficient markets theory and suggest that in the period under review, the Czech stock market was not effective in any form.

scholarly journals Stock Market Predictor using Long short-term Memory (LSTM) Technique

2020 ◽  
Vol 9 (4) ◽  
pp. 393-396
Keyword(s):  
Machine Learning ◽  
Stock Market ◽  
Stock Price ◽  
Short Term Memory ◽  
Learning Model ◽  
Short Term ◽  
Term Memory ◽  
Stock Price Prediction ◽  
Machine Learning Model ◽  
Long Short Term Memory

In the stock market, it is important to have accurate prediction of future behavior of stock price..Because of the great chance of financial loss as well as scoring profits at the same time, it is mandatory to have a secure prediction of the values of the stocks. But when it comes to predicting the value of a stock in future we tend to follow stock market experts but as technology is progressing we may use these technologies rather than following human experts who may be biased many times. Stock price prediction has been interesting area for investors and researchers. This article proposes an approach towards prediction of stock price using machine learning model Long Short Term Memory. This is an ensemble learning method that has been an exceedingly successful model for predicting sequence of numbers and words. Long Short Term Memory is a machine learning model for prediction. This technique is used to forecast the future stock price of a specific stock by using historical data of the stock gathered from Yahoo! Finance.

scholarly journals Evolutionary Framework with Bidirectional Long Short-Term Memory Network for Stock Price Prediction

2021 ◽  
Vol 2021 ◽  
pp. 1-8
Author(s):  
Hongying Zheng ◽  
Hongyu Wang ◽  
Jianyong Chen
Keyword(s):  
Stock Market ◽  
Stock Prices ◽  
Stock Price ◽  
Short Term Memory ◽  
Optimal Solution ◽  
Short Term ◽  
Term Memory ◽  
Stock Price Prediction ◽  
Memory Network ◽  
Long Short Term Memory

As an important part of the social economy, stock market plays an important role in economic development, and accurate prediction of stock price is important as it can lower the risk of investment decision-making. However, the task of predicting future stock price is very difficult. This difficulty arises from stocks with nonstationary behavior and without any explicit form. In this paper, we propose a novel bidirectional Long Short-Term Memory Network (BiLSTM) framework called evolutionary BiLSTM (EBiLSTM) for the prediction of stock price. In the framework, three independent BiLSTMs correspond to different objective functions and act as mutation individuals, then their respective losses for evolution are calculated, and finally, the optimal objective function is identified by the minimum of loss. Since BiLSTM is effective in the prediction of time series and the evolutionary framework can get an optimal solution for multiple objectives, their combination well adapts to the nonstationary behavior of stock prices. Experiments on several stock market indexes demonstrate that EBiLSTM can achieve better prediction performance than others without the evolutionary operator.

Forecasting based on spectral time series analysis: prediction of the Aurubis stock price

2020 ◽  
Vol 17 (4) ◽  
pp. 215-227
Author(s):  
Julia Babirath ◽  
Karel Malec ◽  
Rainer Schmitl ◽  
Kamil Maitah ◽  
Mansoor Maitah
Keyword(s):  
Time Series ◽  
Risk Management ◽  
Time Series Analysis ◽  
Wiener Process ◽  
Stock Price ◽  
Percentage Error ◽  
Portfolio Performance ◽  
Stock Price Prediction ◽  
Series Analysis ◽  
Augmented Dickey Fuller

The attempt to predict stock price movements has occupied investors ever since. Reliable forecasts are a basis for investment management, and improved forecasting results lead to enhanced portfolio performance and sound risk management. While forecasting using the Wiener process has received great attention in the literature, spectral time series analysis has been disregarded in this respect. The paper’s main objective is to evaluate whether spectral time series analysis can produce reliable forecasts of the Aurubis stock price. Aurubis poses a suitable candidate for an investor’s portfolio due to its sound economic and financial situation and the steady dividend policy. Additionally, reliable management contributes to making Aurubis an investment opportunity. To judge if the achieved forecast results can be considered satisfactory, they are compared against the simulation results of a Wiener process. After de-trending the time series using an Augmented Dickey-Fuller test, the residuals were compartmentalized into sine and cosine functions. The frequencies, amplitude, and phase were obtained using the Fast Fourier transform. The mean absolute percentage error measured the accuracy of the stock price prediction, and the results showed that the spectral analysis was able to deliver superior results when comparing the simulation using a Wiener process. Hence, spectral time series can enhance stock price forecasts and consequently improve risk management.

scholarly journals A graph-based CNN-LSTM stock price prediction algorithm with leading indicators

2021 ◽  
Author(s):  
Jimmy Ming-Tai Wu ◽  
Zhongcui Li ◽  
Norbert Herencsar ◽  
Bay Vo ◽  
Jerry Chun-Wei Lin
Keyword(s):  
Neural Network ◽  
Stock Market ◽  
Financial Management ◽  
Stock Price ◽  
Short Term Memory ◽  
Prediction Algorithm ◽  
Leading Indicators ◽  
Stock Price Prediction ◽  
Price Prediction ◽  
Wealth Management

AbstractIn today’s society, investment wealth management has become a mainstream of the contemporary era. Investment wealth management refers to the use of funds by investors to arrange funds reasonably, for example, savings, bank financial products, bonds, stocks, commodity spots, real estate, gold, art, and many others. Wealth management tools manage and assign families, individuals, enterprises, and institutions to achieve the purpose of increasing and maintaining value to accelerate asset growth. Among them, in investment and financial management, people’s favorite product of investment often stocks, because the stock market has great advantages and charm, especially compared with other investment methods. More and more scholars have developed methods of prediction from multiple angles for the stock market. According to the feature of financial time series and the task of price prediction, this article proposes a new framework structure to achieve a more accurate prediction of the stock price, which combines Convolution Neural Network (CNN) and Long–Short-Term Memory Neural Network (LSTM). This new method is aptly named stock sequence array convolutional LSTM (SACLSTM). It constructs a sequence array of historical data and its leading indicators (options and futures), and uses the array as the input image of the CNN framework, and extracts certain feature vectors through the convolutional layer and the layer of pooling, and as the input vector of LSTM, and takes ten stocks in U.S.A and Taiwan as the experimental data. Compared with previous methods, the prediction performance of the proposed algorithm in this article leads to better results when compared directly.

Classification of Technological Development Trends Based on the Analysis of US Patent Time Series. An Empirical Approach

2021 ◽  
pp. 251-278
Author(s):  
Vladimir I. Karnyshev ◽  
◽  
Vladimir I. Avdzeyko ◽  
Evgenia S. Paskal ◽  
◽  
...  
Keyword(s):  
Time Series ◽  
New Technologies ◽  
Practical Interest ◽  
The United States ◽  
Direct Access ◽  
Investment Planning ◽  
Short Term ◽  
Development Trends ◽  
Technical Problems

The forecasting of development trends and the timely revealing of new technical (technological) fields are the key prerequisite for an effective development of modern economy. Only reliable results of technological analysis (forecast) allow identifying new technologies, understanding the evolution of entire industries, carrying out strategic investment planning at the state level, and also planning R&D correctly. The aim of this work is to justify one of the possible approaches to the classification of technical (technological) fields in terms of assessing their relevance, novelty and short-term prospects. This approach is based on patent analysis, in particular, on the study of the time series features of US invention patents (1976-2018) for more than seventy-three thousand main groups (subgroups) of the 17th edition of the International Patent Classification (IPC17). The United States Patent and Trademark Office (USPTO) has been selected as the primary source of information because it is one of the world’s largest and constantly updated patent resources, providing direct access to full-text descriptions. In the authors’ opinion, a feature analysis of the US patent issue dynamics at time intervals (1976-2015, 2009-2018 and 2016-2018) allows dividing the IPC groups (subgroups) into the following three main clusters: “unpromising”, “promising” and “breakthrough”. In terms of the timely revealing of new, previously unknown, technologies or solutions in the technical field, or of the steadily growing technological trends, the “breakthrough” and “promising” subgroups are of the greatest practical interest. The article presents the results of an empirical classification of 71,266 subgroups (with a non-zero number of the issued patents since 1976 to 2018) in eight sections of the IPC17. These data may be useful for developers, researchers and R&D planners in solving complex scientific and technical problems, as well as for making short-term forecast estimates of a specific technical (technological) field development.

Sign in / Sign up

Export Citation Format

Share Document