Locally weighted regression for sunspots estimation and prediction

Locally weighted regression (LOESS) is a modern non-parametric method designed for treating cases where classical procedures are not highly efficient or cannot applied efficiently. Sunspots the darker areas of solar sphere's surface relative to other regions and an important indicator activity .The aim this paper model predict number sunspots because their very importance understanding terrestrial consequences its direct impact on weather communication systems Earth, which may lead damage satellites. In paper, represented by annual data period from 1900 2021 (122 years) as well monthly January 2022 (1465 months) was obtained global center (Sunspot Index Long-term Solar Observations) (SILSO). The LOESS used estimating predicting sunspots. smoothing parameter, degree polynomial that fulfills lowest Akaike corrected information criterion. analysis showed ability represent sunspot passing diagnostic tests high predictive ability. From values data, it found maximum average will be 123.7 in July 2022, February with 61.3 Regarding year 2023 161.7 sunspots, 2029 16.1. Keywords: regression; sunspot; cycle; prediction.