AMIR KHALEEL HASSOO (amir.hasoo@soran.edu.iq)
Soran University Department of Statistics
ABSTRACT
Forecasting agricultural production is crucial for strategic planning and policy-making. This study employs the Bayesian Structural Time Series (BSTS) model to forecast maize production in Romania for the period 2023-2027. The BSTS model, known for its flexibility and ability to incorporate multiple components like trends, seasonality, and regression effects, is particularly suitable for capturing the complex dynamics of agricultural time series data. Historical data on maize production from 1961 to 2022 in FAOSTAT website was used to train the model, ensuring robust and accurate forecasts. The results indicate a steady increase in maize production over the forecast period, with projected figures of 11,341,460 metric tons in 2023, rising to 11,437,732 metric tons in 2024, 11,558,277 metric tons in 2025, 11,594,832 metric tons in 2026, and 11,578,402 metric tons in 2027. These forecasts provide valuable insights for policymakers, farmers, and stakeholders in the agricultural sector, enabling them to make informed decisions regarding resource allocation, market strategies, and food security planning. The study highlights the efficacy of the BSTS model in agricultural forecasting and underscores its potential application in other areas of economic and environmental planning. Future research could enhance the model by incorporating additional variables such as climate data and economic indicators, further improving the accuracy and reliability of agricultural forecasts.
Keywords: BSTS Model, Maize Production Forecasting, Agricultural Planning, Romania.
