Published Online:July 2025
Product Name:The IUP Journal of Applied Economics
Product Type:Article
Product Code:IJAE010725
DOI:10.71329/IUPJAE/2025.24.3.5-23
Author Name:Vengalarao Pachava, Shashi Kumar R and Bhargav Rohit Bolla
Availability:YES
Subject/Domain:Economics
Download Format:PDF
Pages:5-23
The dynamic evolution and significant volatility of cryptocurrency markets create distinct challenges and possibilities for predictive modeling. This study responds to the pressing demand for accurate forecasting techniques capable of handling the unpredictable price movements characteristic of digital assets. It centers on comparing and assessing a range of sophisticated deep learning models, namely, gated recurrent units (GRU), long short-term memory (LSTM), recurrent neural networks (RNN), and multilayer perceptron (MLP) to identify the optimal model for forecasting cryptocurrency prices. The study utilizes daily closing price data for Bitcoin, Binance Coin, Ether, Maker, and Solana across a three-year span (2021- 2024). Each model’s performance is systematically evaluated using five essential error metrics: mean absolute error (MAE), mean absolute percentage error (MAPE), mean squared error (MSE), root mean squared error (RMSE), and symmetric mean absolute percentage error (SMAPE). The findings highlight the GRU model’s consistent superiority, with lower error rates and an enhanced capacity to capture intricate sequential patterns, indicating its effectiveness for forecasting within volatile cryptocurrency contexts. These results emphasize GRU’s potential as a leading model for cryptocurrency price prediction, with promising applications in hybrid or ensemble configurations to further strengthen forecasting accuracy
The advent of cryptocurrencies has introduced an innovative financial ecosystem, marked by decentralized digital assets that operate without the oversight of traditional financial institutions. Since the inception of Bitcoin in 2009, the cryptocurrency market has rapidly evolved, encompassing thousands of digital currencies, each exhibiting unique characteristics