Modeling and Forecasting Cryptocurrency Closing Prices with Rao Algorithm-Based Artificial Neural Networks: A Machine Learning Approach

Abstract

Artificial neural networks (ANNs) are suitable procedures for predicting financial time series (FTS). Cryptocurrencies are good investment assets; therefore, the effective prediction of cryptocurrencies has become a trending area of research. Capturing inherent uncertainties associated with cryptocurrency FTS with conventional methods is difficult. Though ANNs are the better alternative, fixing the optimal parameters of ANNs is a tedious job. This article develops a hybrid ANN through Rao algorithm (RA + ANN) for the effective prediction of six popular cryptocurrencies such as Bitcoin, Litecoin, Ethereum, CMC 200, Tether, and Ripple. Six comparative models such as GA + ANN, PSO + ANN, MLP, SVM, LSE, and ARIMA are developed and trained in a similar way. All these models are evaluated through the mean absolute percentage of error (MAPE) and average relative variance (ARV) metrics. It is found that the proposed RA + ANN generated the lowest MAPE and ARV values, statistically different as compared with existing methods mentioned above, and hence can be recommended as a potential financial instrument for predicting cryptocurrencies.

Minimum, Maximum, and Average Implementation of Patterned Datasets in Mapping Cryptocurrency Fluctuation Patterns

“…For example, comparative analysis research uses random forest models to predict Bitcoin [21], [7]. Patterned datasets can also be used as additional features in research predicting the next bitcoin price using machine learning techniques [13], combining machine learning models, SARIMA, and Facebook Prophet [16], forecasting cryptocurrency returns [29], Rao's algorithm-based artificial neural network [27], and other similar forms of research. Fuzzy prediction models, such as those used in computational intelligence engineering research using hybrid neuro-fuzzy controllers [14], QFTS models and QFTS-ANN hybrids [26], and dataset-based fuzzy modeling [18], can also use patterned datasets as data components to predict the direction of changes in cryptocurrency prices.…”

Artificial intelligence applications in solar energy

“…4 that the fundamental structure of the ANN comprises three levels: the input layer, the hidden layer, and the output layer. Each of these layers is composed of neurons [128]. Whereas the selection of input parameters determines the number of neurons in the input layer, the selection of output parameters determines the number of neurons in the output layer [129], [130].…”

“…Fig. 4 ANN architecture [128] During the process of training a neural network, several different training algorithms are used. The training function, the learning variant, the transfer function, and the number of hidden neurons are all taken into consideration by these respective approaches [142].…”

Untitled

“…The industries now associated with FinTech include education, retail banking, fundraising, non-profits, and investment companies [44]. Each of these companies and industries now achieves a specific type of growth by ensuring financing effectiveness [45]. The overall applications of this type of development have applied trust across the finance sectors of these organizations.…”