Modeling steel supply and demand functions using logarithmic multiple regression analysis (case study: Steel industry in Iran)

Abstract

The steel industry is considered as one of the mother industries that serves many large and small industries. Hence, recognizing the market situation of the steel industry both inside and outside the country is very important. Supply and demand are among the most important factors in stimulating the steel market. Usually, supply and demand have complex function. So, modeling and forecasting steel supply and demand require the use of accurate and scientific approaches. This paper presents an approach to identify the steel supply and demand functions and also to forecast the supply and demand trends. In the first step, through reviewing the historical data on the steel supply and demand in Iran, the effective and most important variables will be identified. Then, the supply and demand functions will be fitted using multiple logarithmic regression analysis. Logarithmically transforming variables in a regression model is a very common way to handle situations where a non-linear relationship exists between the independent and dependent variables. The accuracy of estimations is checked through appropriate statistical tests. The analysis is based on data of Iran steel market obtained from a 60 monthly period starting in 2010 and ending in 2014. The results showed that the estimated functions was appropriate in modeling the steel supply and demand behavior. The extrapolation analysis using 24 monthly data from 2017-2018 has also been accomplished to check the performance of the regression analysis.