Application of the Double Exponential Smoothing Brown Method in Forecasting the Number of Poor Population

Intan Utami
Ana Istiqomah
Sangidatus Sholiha

Sciencestatistics: Journal of Statistics, Probability, and Its Application
Vol. 4 No. 1 (2026): JANUARY

Poverty is a socio-economic issue that requires policy planning based on accurate forecasting. This study uses the Brown Double Exponential Smoothing method to forecast the number of poor people in Metro City for the period 2026-2030, using data from 2005-2025 obtained from BPS. The analysis was conducted using a trial and error method for the alpha (α) parameter ranging from 0.1 to 0.9 based on the smallest MAD, MSE, and MAPE values. The study results indicate that the optimal alpha parameter is α = 0.5 with a MAPE of 15.91231%, which indicates good accuracy. The forecast shows an increasing trend from 321.90 thousand people (2026) to 338.87 thousand people (2030), with an average increase of 4.24 thousand people per year. The results of this study can be used as a basis for planning poverty alleviation programs in Metro City.