Forecasting Cotton Production in Iraq during the years (1960-2022) using Markov Chain Approach and Holt-Winter Method

Abstract

Cotton is a fibrous material derived from the seed pods of the cotton plant (Gossypium). It is a natural fiber extensively utilized in the textile industry for the manufacturing of items like clothing, linens, and other fabric-based products. Valued for its breathability, absorbency, and adaptability, cotton is a widely chosen material for diverse everyday goods. Two models are used in this study, such as the Markov chain approach and the Holt-Winter method, to forecast cotton production in Iraq over the years 1960–2022. A Markov chain approach model is a accurate framework describing a series of states in a system. The chance of moving from one state to another depends only on the present state, without consideration of the historical. This model adheres to the Markov property, exhibiting a memoryless characteristic. It encompasses a set of states, transition probabilities between these states, and a stochastic process evolving over discrete time intervals. The Holt-Winters method is a robust technique for forecasting time series data, particularly when the data exhibits both trend and seasonality. This method integrates three key components into its forecasting model: level , trend  and seasonality . The data for this study was obtained from the website: https://www.indexmundi.com/agriculture. The study evaluates the performance of the two forecasting models. The results show that the Holt-Winter method is more accurate than the Markov chain dependent on RMSE, MAE, and MAPE, and cotton production in Iraq will decrease over the coming years.