Perbandingan Moving Average dan Exponential Smoothing untuk Prediksi Harga Saham BBRI pada Dataset 2019–2026

Wahyu Dedy Setiyawan (Universitas PGRI Ronggolawe, Tuban)
Andy Haryoko (Universitas PGRI Ronggolawe, Tuban)
Amaludin Arifia (Universitas PGRI Ronggolawe, Tuban)

JURIKOM (Jurnal Riset Komputer)
Vol. 13 No. 3 (2026): Juni 2026

This study compares four time series forecasting methods Simple Moving Average (SMA), Double Moving Average (DMA), Single Exponential Smoothing (SES), and Double Exponential Smoothing (DES/Holt) for predicting the closing stock price of BBRI.JK. The dataset comprises 1,768 daily observations spanning January 2019 to December 2026, split into training (80%) and testing (20%) sets. Each method's parameters were optimized via grid search minimizing MAPE, then evaluated across three metrics: MAPE, MAE, and RMSE. SES (α = 0.9) emerged as the best-performing model, achieving a MAPE of 0.3763%, MAE of IDR 14.93, and RMSE of IDR 24.31 substantially outperforming SMA (3.1591%), DMA (2.7561%), and DES (3.6973%). These findings offer methodological guidance for researchers and practical insight for investors operating in emerging market equities with near weak-form efficiency.