Work

Computational Framework Capabilities to Improve Export Forecasting

Year: 2026

Type: conference-paper

Abstract: In this article, we review how we can use the Python computational framework as a complement to traditional economics statistical strategies to efficiently manage exporting companies' forecasts. We collect information over a 21-year period to analyze the influence of certain macroeconomic indicators on the forecasting of product exports. The Python™ Framework is utilized to generate short-term forecasts for two Ecuadorian export products (shrimp and banana) and to evaluate the effect of 3 macroeconomic indicators (EMBI, FDI and exchange rate) with an ARMAX model. Among the main findings, we identified that the proposed methodology is not very effective for the shrimp case, but is quite accurate when evaluating banana exports, where correlation coefficients of up to 0.97 were reached.

Source: International Conference on Electrical, Computer and Energy Technologies (ICECET)

DOI: 10.1109/icecet65726.2026.11632917

Authors: R. Llugsi, D. Salgado, R. Álvarez, C. Tipantuña, P. Lupera

Institution: National Polytechnic School

Language: English

Topic: Forecasting Techniques and Applications

Subfield: Management Science and Operations Research

Field: Decision Sciences

Domain: Social Sciences

SDGs: Industry, innovation and infrastructure, Decent work and economic growth

Open Access status: hybrid

OpenAlex ID: W7202299111

FWCI:

Cites: 12

Cited by:

Related to: 10