Comparison of singular spectrum analysis forecasting algorithms for student’s academic performance during COVID-19 outbreak

Due to the spread of COVID-19 that hit Malaysia, all academic activities at educational institutions including universities had be carried out via online learning However, effectiveness is remains unanswered Besides, may have a significant impact if continued in upcoming sessions Therefore, core this study predict performance undergraduate students one public Malaysia by using Recurrent Forecasting-Singular Spectrum Analysis (RF-SSA) and Vector (VF-SSA) The key concept predictive model improve efficiency different types forecast SSA two parameters which are window length (L) number leading components (r) forecasting approaches was based on Grading Point Assessments (GPA) for from Faculty Science Mathematics, UPSI classes during outbreak experiment revealed parameter L= 11 (T/20) has best prediction result RF-SSA with RMSE value 0 19 as compared VF-SSA 30 This signifies competency predicting students’ performances GPA semester Nonetheless, an algorithm should developed higher effectivity obtaining more data sets respondents various © 2021 Muhammad Fakhrullah Mohd Fuad et al