<![CDATA[The increased intensity of rainfall is becoming one of the most pressing climate-related issues in many parts of the world. Detecting the factors that affect rainfall intensity requires a combination of modern technologies, such as weather satellites, radar systems, and advanced atmospheric models. Extreme conditions (outliers) often occur. This study aims to model data that is not symmetric or contains outliers. This study examines and models quantile regression on daily rainfall intensity in Jakarta which has extreme rainfall events. The results of the study found that the extreme values in the daily rainfall intensity data in Jakarta are outliers and the assumptions on modeling using linear regression are not satisfied so that the characteristics of the parameter estimator based on OLS do not have BLUE characteristic. In modeling with quantile regression using six quantiles 0.25, 0.50, 0.75, 0.95, 0.99, and 0.9999 with consideration of these quantile values representing all parts of the data distribution including extreme values, it was found that the factors affecting rainfall intensity in Jakarta are different in each rainfall intensity condition. The best model is shown by quantile 0.999 with a coefficient of determination of 58.21%. Based on the best model, it is known that the factors affecting extreme rainfall are maximum temperature, dew point temperature, air humidity, wind speed, air pressure, and length of irradiation. This study indicates that quantile regression can provide a more detailed insight into how these variables affect rainfall intensity in various rainfall conditions ranging from low rainfall to extreme rainfall.]]>
Copyrights © 2025
