Implementasi metode double exponential smoothing pada harga komoditas nikel future

Abstract

Nickel is a strategic commodity in the stainless steel and electric vehicle battery industries, which experience high price volatility. This study aims to optimize future historical nickel price predictions by applying the Double Exponential Smoothing (DES) method, which is capable of capturing trend patterns through both level and trend components simultaneously. The data used are daily nickel closing prices for the period January 1, 2020, to August 21, 2025, sourced from Investing.com. The research methodology involves dividing the data into training data (2020–2023) for model development and test data (2024) for accuracy validation. The implementation was carried out using the Python programming language using a time series data analysis library. The test results show that the DES model produces a very high level of accuracy with a Mean Absolute Percentage Error (MAPE) value of 1.23%. This achievement demonstrates that the DES method is highly effective and reliable as a supporting instrument for strategic decision-making in the mining commodity investment sector.