Research on a hybrid prediction model for stock price based on long short-term memory and variational mode decomposition

Yang, Yujun; Yang, Y.B.; Wang, Zhou

doi:10.1007/s00500-021-06122-4

“…The lower the forecast error, the lower the investment risk and the higher the profitability [4]. The formation and fluctuation of stock prices are not only subject to various economic and political factors but are also influenced by investment psychology and trading techniques [5]. For investors, there is a direct link between the prediction of stock market trends and the acquisition of profits, and the more accurate the prediction, the more certain the prevention of risks [5].…”

Section: Introductionmentioning

confidence: 99%

“…The formation and fluctuation of stock prices are not only subject to various economic and political factors but are also influenced by investment psychology and trading techniques [5]. For investors, there is a direct link between the prediction of stock market trends and the acquisition of profits, and the more accurate the prediction, the more certain the prevention of risks [5]. For listed companies, the stock index reflects the company's operations and future trends and is the main technical indicator for analyzing the company [5,6].…”

Section: Introductionmentioning

confidence: 99%

“…For investors, there is a direct link between the prediction of stock market trends and the acquisition of profits, and the more accurate the prediction, the more certain the prevention of risks [5]. For listed companies, the stock index reflects the company's operations and future trends and is the main technical indicator for analyzing the company [5,6].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Stock Price Prediction based on Multiple Regression Models

Li¹

2023

HSET

View full text Add to dashboard Cite

Over the past two years, global stock markets have gradually recovered and new investors have entered the market. While there are many factors affecting stock prices and the stock market is changing rapidly, the way to accurately predict stock prices has become the focus of investors. This paper will use the concept of machine learning to predict the stock prices of three listed companies based on three different regression models (i.e., OLS, Ridge and XGBoost). According to the analysis, the OLS model and the Ridge model are very accurate in predicting stock prices, especially in the low and middle price ranges. In contrast to these typical linear regression models, the XGBoost model is not as accurate in predicting stock prices and even has a significant prediction bias in the high price range. These results will enable subsequent research to make better choices when selecting models for forecasting, especially for data sets with different characteristics.

show abstract

“…These band-limited components reconstruct the input time series exactly or in the least squares sense. In other words, VMD is capable of separating tones of similar frequencies and avoiding the mode mixing problem [1,10,15].…”

Section: Introductionmentioning

confidence: 99%

Multi-step-ahead Stock Price Prediction Using Recurrent Fuzzy Neural Network and Variational Mode Decomposition

Nasiri¹,

Ebadzadeh²

2022

Preprint

5

0

View full text Add to dashboard Cite

<p>Financial time series prediction, a growing research topic, has attracted considerable interest from scholars, and several approaches have been developed. Among them, decomposition-based methods have achieved promising results. Most decomposition-based methods approximate a single function, which is insufficient for obtaining accurate results. Moreover, most existing researches have concentrated on one-step-ahead forecasting that prevents stock market investors from arriving at the best decisions for the future. This study proposes two novel methods for multi-step-ahead stock price prediction based on the issues outlined. DCT-MFRFNN, a method based on discrete cosine transform (DCT) and multi-functional recurrent fuzzy neural network (MFRFNN), uses DCT to reduce fluctuations in the time series and simplify its structure and MFRFNN to predict the stock price. VMD-MFRFNN, an approach based on variational mode decomposition (VMD) and MFRFNN, brings together their advantages. VMD-MFRFNN consists of two phases. The input signal is decomposed to several IMFs using VMD in the decomposition phase. In the prediction and reconstruction phase, each of the IMFs is given to a separate MFRFNN for prediction, and predicted signals are summed to reconstruct the output. Three financial time series, including Hang Seng Index (HSI), Shanghai Stock Exchange (SSE), and Standard & Poor's 500 Index (SPX), are used for the evaluation of the proposed methods. Experimental results indicate that VMD-MFRFNN surpasses other state-of-the-art methods. VMD-MFRFNN, on average, shows 35.93%, 24.88%, and 34.59% decreases in RMSE from the second-best model for HSI, SSE, and SPX, respectively. Also, DCT-MFRFNN outperforms MFRFNN in all experiments.</p>

show abstract

“…Another issue that requires attention is that financial time series are usually dynamic, non-linear, extremely fluctuated, and complex, which makes their prediction challenging [9][10][11]. To address this issue, decomposition-based methods have been proposed.…”

Section: Introductionmentioning

confidence: 99%

Multi-step-ahead Stock Price Prediction Using Recurrent Fuzzy Neural Network and Variational Mode Decomposition

Nasiri¹,

Ebadzadeh²

2022

Preprint

1

0

View full text Add to dashboard Cite

<p>Financial time series prediction, a growing research topic, has attracted considerable interest from scholars, and several approaches have been developed. Among them, decomposition-based methods have achieved promising results. Most decomposition-based methods approximate a single function, which is insufficient for obtaining accurate results. Moreover, most existing researches have concentrated on one-step-ahead forecasting that prevents stock market investors from arriving at the best decisions for the future. This study proposes two novel methods for multi-step-ahead stock price prediction based on the issues outlined. DCT-MFRFNN, a method based on discrete cosine transform (DCT) and multi-functional recurrent fuzzy neural network (MFRFNN), uses DCT to reduce fluctuations in the time series and simplify its structure and MFRFNN to predict the stock price. VMD-MFRFNN, an approach based on variational mode decomposition (VMD) and MFRFNN, brings together their advantages. VMD-MFRFNN consists of two phases. The input signal is decomposed to several IMFs using VMD in the decomposition phase. In the prediction and reconstruction phase, each of the IMFs is given to a separate MFRFNN for prediction, and predicted signals are summed to reconstruct the output. Three financial time series, including Hang Seng Index (HSI), Shanghai Stock Exchange (SSE), and Standard & Poor's 500 Index (SPX), are used for the evaluation of the proposed methods. Experimental results indicate that VMD-MFRFNN surpasses other state-of-the-art methods. VMD-MFRFNN, on average, shows 35.93%, 24.88%, and 34.59% decreases in RMSE from the second-best model for HSI, SSE, and SPX, respectively. Also, DCT-MFRFNN outperforms MFRFNN in all experiments.</p>

show abstract

Research on a hybrid prediction model for stock price based on long short-term memory and variational mode decomposition

Cited by 27 publications

References 67 publications

Stock Price Prediction based on Multiple Regression Models

Stock Price Prediction based on Multiple Regression Models

Multi-step-ahead Stock Price Prediction Using Recurrent Fuzzy Neural Network and Variational Mode Decomposition

Multi-step-ahead Stock Price Prediction Using Recurrent Fuzzy Neural Network and Variational Mode Decomposition

Contact Info

Product

Resources

About