On RBL-STEM Learning Activities: The Use of ANN Multi-Step Time Series Forecasting to Improve the Students Metacognition in Solving A Cryptocurency Volatility Based on the Fundamental, Technical and On-Chain Analysis
Metacognitive skills refer to an individual’s ability to be aware of, manage, and evaluate their own thinking processes, supporting independent learning strategies and understanding. However, the low level of metacognitive skills highlights the need for further research, particularly on how to improve them. This study aims to enhance students’ metacognitive skills through a Research-Based Learning (RBL) approach and STEM learning activities by employing multi-step time-series forecasting techniques based on Artificial Neural Networks (ANN). The main focus is on designing learning activities that teach ANN Multi-Step Time Series Forecasting to address cryptocurrency volatility, analyzed using three main approaches: fundamental analysis, technical analysis, and on-chain analysis. By applying ANN to predict cryptocurrency price volatility, students are expected to better understand market dynamics and improve their decision-making abilities. The results show that RBL-STEM learning activities can be structured into six clear stages to utilize ANN Multi-Step Time Series Forecasting for improving students’ metacognitive skills in solving cryptocurrency volatility issues based on these analytical approaches.