Performance of AICC Statistic as Time Series Modeling and Forecasting of Sunspots
International Journal of Artificial Intelligence & Mathematical Sciences · 2023
DOI: 10.58921/ijaims.v1i2.38 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Artificial Intelligence & Mathematical Sciences, 2023. Verfasst von Ali Khan und 5 weiteren Autor:innen.
Abstract (Originalsprache)
Dynamics of the Sun is quite complex. Sunspot number is an important index to understand the solar dynamo mechanism that governs the solar magnetic cycle. This paper is an attempt to forecast Sunspots number by finding an appropriate time series model. For the purpose, Sunspots datasets from 1749 to 2010 were used. Also, 23 minimum to minimum (m-m) and 24 maximum to minimum (M-m) Sunspots cycles were investigated. The results show that Autoregressive Moving average (ARMA) time series models fit the Sunspots number well. They have corrected predictions of the future trend of the Sunspots number within the sample period of study. The results obtained showed that the probability of the AIC corrected model was found better fitted. The problem of over parameterization exited in the model used, but under parameterization was found to minimal.
Zitieren
APA
Khan, A., Murshid, S. M., Anwer, A., Ali, A., Afrahim, S., & Zafar, M. (2023). Performance of AICC Statistic as Time Series Modeling and Forecasting of Sunspots. International Journal of Artificial Intelligence & Mathematical Sciences. https://doi.org/10.58921/ijaims.v1i2.38
Vancouver
Khan A, Murshid SM, Anwer A, Ali A, Afrahim S, Zafar M. Performance of AICC Statistic as Time Series Modeling and Forecasting of Sunspots. International Journal of Artificial Intelligence & Mathematical Sciences. 2023. doi:10.58921/ijaims.v1i2.38
BibTeX
@article{Khan2023,
author = {Khan, Ali and Murshid, Syed Muhammad and Anwer, Afshan and Ali, Amjad and Afrahim, Sobia and Zafar, Mahreen},
title = {Performance of AICC Statistic as Time Series Modeling and Forecasting of Sunspots},
journal = {International Journal of Artificial Intelligence & Mathematical Sciences},
year = {2023},
doi = {10.58921/ijaims.v1i2.38}
}
EUVIMED