Nonparametric Model For Tin Commodity Price Prediction In Bangka Belitung Using Time Series Analysis

Abstract

Tin commodity prices are highly volatile and influenced by various global and local factors. The nonparametric model in time series analysis offers a more flexible approach than the parametric model in capturing patterns of price fluctuations. This study aims to build a tin price prediction model in Bangka Belitung using a nonparametric regression method with a smoothing spline and kernel regression approach. Tin price data collected from official sources is analyzed using time series-based nonparametric methods to get more accurate predictions. The results show that the nonparametric model has a better performance than the conventional model in capturing the pattern of tin price changes. These findings can provide benefits for industry players and policymakers in anticipating fluctuations in commodity prices.