WiMi Developed a Hybrid Machine Learning Model Based on VMD and SVR to Lead Bitcoin Price Prediction
VMD is able to better handle noise and random fluctuations in Bitcoin price series. By decomposing the real-valued input signals into variational mode function (VMF), we obtain VMFs with unique frequency ranges, which ultimately improves the representation of price data. SVR, a core component of the machine learning algorithms, provides powerful predictive capabilities by capturing nonlinear relationships in the feature space of the technical model. The hybrid input of technical indicators and the reconstructed VFMs of the VMD allow SVR to provide a more comprehensive understanding of market dynamics. To ensure the relevance of the predictive model data, intraday bitcoin price data was preprocessed and normalized. This included converting heterogeneous time series data to homogeneous data to eliminate differences in scale, thus making support vectors easier to learn.
Firstly, in the first stage, the Boruta algorithm, which is an efficient feature selection algorithm, is employed to select the most relevant subset from various technical metrics. The purpose of this step is to reduce the feature space and decrease the complexity of the model while ensuring that the selected technical indicators are maximally informative for Bitcoin price prediction.
The VMD then decomposes the Bitcoin price series into a set of VMFs. Each VMF has unique properties and frequency ranges, allowing us to more accurately capture noisy signals and random fluctuations in the price data. This step results in a reconstructed set of variational modal functions (rVMFs), which provide cleaner and more abstract inputs for the second stage of modeling.
In the second stage, information from two feature sets is aggregated to form the inputs to the SVR. These two feature sets include features selected through technical indicators and rVMFs generated through VMDs. This aggregation is designed to fully utilize the statistical trends of the technical indicators and the frequency information of the VMDs to provide a more comprehensive, multidimensional input to SVR.
SVR is the core of the model and has the ability to capture non-linear relationships. Accepting a mixture of inputs from both feature sets, SVR builds a powerful predictive model by learning from past market behavior and statistical patterns of price movements. Since this model takes into account both technical indicators and frequency domain information from VMDs, it provides a more comprehensive understanding of the volatility of the Bitcoin price.
Through two-stage hybrid modeling, WiMi combines the statistical properties of technical indicators with the frequency domain information of VMDs to construct a more comprehensive and powerful forecasting model. This model demonstrates significant advantages in dealing with market volatility, handling noise, and adapting to rapid changes. It not only improves the accuracy of Bitcoin price forecasts, but also provides more actionable decision support.
As the cryptocurrency market continues to evolve and innovate, the need for technology continues to escalate. Going forward, WiMi will continue to deepen its market data and integrate more emerging technologies to further enhance the performance of its two-stage hybrid machine learning model. By planning to introduce more advanced machine learning algorithms, augmented learning methods, and deep learning techniques to adapt to the dynamic changes in the market, WiMi will provide users with more accurate and reliable Bitcoin price predictions.
In the digital asset space, WiMi's two-stage hybrid machine-learning model marks a technology innovation. Through in-depth research of the Bitcoin market and the application of cutting-edge technology, it breaks the limitations of traditional models and provides investors and traders with a new, more reliable tool for Bitcoin price prediction. WiMi provides an unprecedented approach to bitcoin price prediction. The development of this model is not only an important contribution to the field of financial technology, but also provides investors and traders with a more reliable and efficient decision-support tool.
About
Safe Harbor Statements
This press release contains "forward-looking statements" within the Private Securities Litigation Reform Act of 1995. These forward-looking statements can be identified by terminology such as "will," "expects," "anticipates," "future," "intends," "plans," "believes," "estimates," and similar statements. Statements that are not historical facts, including statements about the Company's beliefs and expectations, are forward-looking statements. Among other things, the business outlook and quotations from management in this press release and the Company's strategic and operational plans contain forward−looking statements. The Company may also make written or oral forward−looking statements in its periodic reports to the
Further information regarding these and other risks is included in the Company's annual report on Form 20-F and the current report on Form 6-K and other documents filed with the
View original content:https://www.prnewswire.com/news-releases/wimi-developed-a-hybrid-machine-learning-model-based-on-vmd-and-svr-to-lead-bitcoin-price-prediction-302110368.html
SOURCE