Backpropagation neural network prediction for cryptocurrency bitcoin prices

Sovia, Rini; Yanto, Musli; Budiman, Arif; Mayola, Liga; Saputra, Dio

doi:10.1088/1742-6596/1339/1/012060

Cited by 7 publications

(7 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another approach is to study the time series and make a prediction based on the processing and analysis of past observations. The most common models are the Box–Jenkins ARIMA time‐series models and their modifications, GARCH models, or ANNs (Alahmari, 2019; Derbentsev et al, 2019; Ho, Xie, & Goh, 2002; Lu, 2010; Radityo et al, 2017; Sovia et al, 2019; Vapnik, 1999). Generally, the models for predicting cryptocurrency prices depend on an analyst's perception of the causal relationships in the pricing process.…”

Section: Discussionmentioning

confidence: 99%

“…So the next best candidate was the BPNN with 300 times faster accuracy time and slightly less accuracy compared with the GABPNN. The price movement of bitcoin was predicted using ANNs based on back‐propagation algorithm in Sovia, Yanto, Budiman, Mayola, and Saputra (2019) using graph movements: open, low, and high, bitcoin requests, volumes, and next hour prediction prices. Prediction variables with target values can be predicted using prior bitcoin price.…”

Section: Techniques For Cryptocurrency Price Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Cryptocurrency price prediction using traditional statistical and machine‐learning techniques: A survey

Khedr

Arif

et al. 2021

Intell Sys Acc Fin Mgmt

View full text Add to dashboard Cite

Cryptocurrencies are decentralized electronic counterparts of government-issued money. The first and best-known cryptocurrency example is bitcoin. Cryptocurrencies are used to make transactions anonymously and securely over the internet. The decentralization behavior of a cryptocurrency has radically reduced central control over them, thereby influencing international trade and relations. Wide fluctuations in cryptocurrency prices motivate the urgent requirement for an accurate model to predict its price. Cryptocurrency price prediction is one of the trending areas among researchers. Research work in this field uses traditional statistical and machine-learning techniques, such as Bayesian regression, logistic regression, linear regression, support vector machine, artificial neural network, deep learning, and reinforcement learning. No seasonal effects exist in cryptocurrency, making it hard to predict using a statistical approach. Traditional statistical methods, although simple to implement and interpret, require a lot of statistical assumptions that could be unrealistic, leaving machine learning as the best technology in this field, being capable of predicting price based on experience. This article provides a comprehensive summary of the previous studies in the field of cryptocurrency price prediction from 2010 to 2020. The discussion presented in this article will help researchers to fill the gap in existing studies and gain more future insight.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Techniques For Cryptocurrency Price Predictionmentioning

confidence: 99%

Cryptocurrency price prediction using traditional statistical and machine‐learning techniques: A survey

Khedr

Arif

et al. 2021

Intell Sys Acc Fin Mgmt

View full text Add to dashboard Cite

show abstract

“…where π θ is a fixed policy; Ê t [• ] represents the empirical average value of a limited batch of samples; a denotes the action and s denotes the state at time t; Â t is an estimator of the dominant function. The estimated value of ĝ is got by differentiation of the objective function, which can be derived as (7):…”

Section: Proximal Policy Optimization (Ppo)mentioning

confidence: 99%

“…References [7] and [8] both build an artificial neural network (ANN) to predict the price of bitcoin, but [8] concentrates on ensemble algorithms for direction prediction, rather than price prediction, which can not give references for high-frequency trading directly. Based on [7], [9] applies the neural network auto-regression (NNAR) to complete the-next-day prediction and finds that NNAR is inferior to ARIMA in daytime prediction, demonstrating that naive neural networks are possibly not useful than traditional methods. With the development of Recurrent Neural Networks (RNNs), the long sequence prediction method has been developed unprecedentedly.…”

Section: Introductionmentioning

confidence: 99%

Bitcoin Transaction Strategy Construction Based on Deep Reinforcement Learning

Liu

Ren

Zhai

et al. 2021

2021 IEEE 2nd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE)

View full text Add to dashboard Cite

The emerging cryptocurrency market has lately received great attention for asset allocation due to its decentralization uniqueness. However, its volatility and brand new trading mode has made it challenging to devising an acceptable automatically-generating strategy. This study proposes a framework for automatic high-frequency bitcoin transactions based on a deep reinforcement learning algorithmproximal policy optimization (PPO). The framework creatively regards the transaction process as actions, returns as awards and prices as states to align with the idea of reinforcement learning. It compares advanced machine learning-based models for static price predictions including support vector machine (SVM), multi-layer perceptron (MLP), long short-term memory (LSTM), temporal convolutional network (TCN), and Transformer by applying them to the real-time bitcoin price and the experimental results demonstrate that LSTM outperforms. Then an automatically-generating transaction strategy is constructed building on PPO with LSTM as the basis to construct the policy. Extensive empirical studies validate that the proposed method perform superiorly to various common trading strategy benchmarks for a single financial product. The approach is able to trade bitcoins in a simulated environment with synchronous data and obtains a 31.67% more return than that of the best benchmark, improving the benchmark by 12.75%. The proposed framework can earn excess returns through both the period of volatility and surge, which opens the door to research on building a single cryptocurrency trading strategy based on deep learning. Visualizations of trading the process show how the model handles high-frequency transactions to provide inspiration and demonstrate that it can be expanded to other financial products.

show abstract

“…Therefore, the fault detection and diagnosis of the batch process have attracted increasing concerns for process safety and quality control. 2−6 In recent years, machine learning methods, such as the back propagation neural network (BPNN) 7 and the support vector machine (SVM), 8 are often used as intelligent control methods in WWTP. However, it is difficult to diagnose and control the SBR operational state due to its complicated and vague mechanism.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Intelligent Fault Diagnosis Model Based on Improved MPCA-V for Sensors in a Laboratory-Scale Wastewater Treatment Process

Liu

Niu

Zhou

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

The study objective is to propose a hybrid fault diagnosis method for a laboratory-scale sequential batch reactor (SBR) wastewater treatment process based on time-varying covariance and variable-wise unfolded MPCA method (MPCA-V), which can detect the fault batch, determine the fault time simultaneously, and further identify the fault source. To establish and validate the MPCA-V model, 50 normal batches and 55 batches including 7 fault batches were employed separately. Furthermore, the classical MPCA (MPCA-B) model was introduced for comparison. For the three detected fault batches, with the MPCA-V model, not only the fault occurring time and fault source were located and identified by the contribution degree calculation of each variable to the T 2 and SPE statistics simultaneously but also the fault detection rate was averaged as 90%, which was much higher than that of MPCA-B (67%). Introducing time dependency and correlation in a laboratory-scale SBR process gives the work practical significance and breakthrough.

show abstract

Backpropagation neural network prediction for cryptocurrency bitcoin prices

Cited by 7 publications

References 3 publications

Cryptocurrency price prediction using traditional statistical and machine‐learning techniques: A survey

Cryptocurrency price prediction using traditional statistical and machine‐learning techniques: A survey

Bitcoin Transaction Strategy Construction Based on Deep Reinforcement Learning

Hybrid Intelligent Fault Diagnosis Model Based on Improved MPCA-V for Sensors in a Laboratory-Scale Wastewater Treatment Process

Contact Info

Product

Resources

About