KEBERLANJUTAN PERDAGANGAN KARET ALAM INDONESIA KE PASAR STRATEGIS: PERBANDINGAN OLS DAN PPML DALAM REGRESI DATA PANEL: The Sustainability of Indonesian Natural Rubber Trade in Strategic Markets: A Comparison of OLS and PPML in Panel Data Regression

Ni Ketut Dela Yanti (Agricultural Economics Study Program, Faculty of Economics and Management, IPB University)
Daffa Ibra Danendra (Agricultural Economics Study Program, Faculty of Economics and Management, IPB University)
Dina Roffida Haqqi Daqianus (Agricultural Economics Study Program, Faculty of Economics and Management, IPB University)
Titin Suhatini Koswara (Agricultural Economics Study Program, Faculty of Economics and Management, IPB University)
Dini Mulia Sari (Agricultural Economics Study Program, Faculty of Economics and Management, IPB University)
Muhamad Alfin Hisam (Agricultural Economics Study Program, Faculty of Economics and Management, IPB University)
Novindra (Department of Resource and Environmental Economics, Faculty of Economics and Management, IPB University, Indonesia)

Indonesian Journal of Agricultural, Resource and Environmental Economics
Vol. 5 No. 1 (2026): June 2026

This study analyzes the determinants of Indonesian natural rubber exports to strategic global markets using Ordinary Least Squares (OLS) and Poisson Pseudo Maximum Likelihood (PPML) on panel data from 2013-2022. The dependent variable is the export value of natural rubber (HS 4001), while the independent variables include real GDP of Indonesia and partner countries, geographical distance, real exchange rates, economic freedom indicators (Government Integrity, Business Freedom, Trade Freedom, Financial Freedom), and infrastructure quality (Container Port Traffic). OLS estimation using the Fixed Effects Model (FEM) reveals that the real GDP of trading partner countries, geographical distance, real exchange rates, and all economic freedom indicators, as well as Container Port Traffic, significantly influence Indonesian rubber exports, with the model explaining 62.48% of the variation in exports. In contrast, PPML only shows the real GDP of trading partner countries as significant, but with a negative coefficient, which is theoretically contradictory. This difference is attributed to PPML's sensitivity to data distribution, particularly the absence of zero trade flows in the export data, leading to an underestimation. Therefore, OLS proves to be more informative in capturing the sensitivity of factors determining Indonesian rubber exports in this data context. Penelitian ini menganalisis faktor-faktor penentu ekspor karet alam Indonesia ke pasar strategis global menggunakan Ordinary Least Squares (OLS) dan Poisson Pseudo Maximum Likelihood (PPML) pada data panel tahun 2013-2022. Variabel dependen adalah nilai ekspor karet alam (HS 4001), sementara variabel independen meliputi PDB riil Indonesia dan negara mitra, jarak geografis, nilai tukar riil, indikator kebebasan ekonomi (Government Integrity, Business Freedom, Trade Freedom, Financial Freedom), dan kualitas infrastruktur (Container Port Traffic). Hasil estimasi OLS dengan Fixed Effect Model (FEM) menunjukkan bahwa PDB riil negara mitra, jarak geografis, nilai tukar riil, serta semua indikator kebebasan ekonomi dan Container Port Traffic secara signifikan memengaruhi ekspor karet Indonesia, dengan model menjelaskan 62,48% variasi ekspor. Sebaliknya, PPML hanya menunjukkan PDB riil negara mitra yang signifikan, namun dengan koefisien negatif yang kontradiktif secara teori. Perbedaan ini disebabkan sensitivitas PPML terhadap distribusi data, khususnya ketiadaan observasi nol (zero trade flows) dalam data ekspor, menyebabkan underestimation. Oleh karena itu, OLS terbukti lebih informatif dalam menangkap sensitivitas faktor-faktor penentu ekspor karet Indonesia dalam konteks data ini.