Eucharia Onyekwere, Francisca Nonyelum Ogwueleka* and Martins Ekata Irhebudhe
Department of Computer Science, Nigerian Defence Academy Kaduna, Nigeria
*Corresponding Author: Francisca Nonyelum Ogwueleka, Department of Computer Science, Nigerian Defence Academy Kaduna, Nigeria.
Received: February 25, 2022; Published: March 25, 2022
Financial time series forecasting could be a daunting task due to inherent noise and non-stationarity. The most crucial and challenging factor in cryptocurrency forecasting is volatility. Cryptocurrency, particularly bitcoin, is not just used to make payments for goods and services; it is traded in exchange for other currencies. In this study, the Multilayer Perceptron (MLP) model with Backpropagation Algorithm written in Python programming language was developed to forecast the volatile time series data of the volume of bitcoin transactions in Nigeria for seven years. Three different models were designed and experimented with varying hidden layers and neurons for the forecast. In comparison, the ever-popular statistical model of Autoregressive Integrated Moving Average (ARIMA) was used as a point of reference for the Neural Network Models. The models were evaluated based on performance measures such as Mean Square Error (MSE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE), which indicated that the best performing MLP model had an MSE of 0.007085, RMSE OF 0.000050 and MAPE of 1.274076. In comparison, the ARIMA model yielded an MSE of 0.092093, RMSE of 0.303468 and MAPE of 46.13.
Keywords: Cryptocurrency; Bitcoin; Blockchain; Artificial Neural Network; Multilayer Perceptron
Citation: Francisca Nonyelum Ogwueleka., et al. “Battery and Solar Panels Temperature Compensation for Small Satellites Applications". Acta Scientific Computer Sciences 4.4 (2022): 72-82.
Copyright: © 2022 Francisca Nonyelum Ogwueleka., et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.