Correlation Between Tractor Variables and Loan Support Limit in South Korea Through Regression Analysis

PURPOSE: The correlation between the major variables of tractors and the loan support limit was investigated through regression analysis. METHODS: The loan support limit according to engine power, weight, engine displacement, width, length, and height was surveyed for 118 tractors commercially available in South Korea. Simple linear regression analysis was performed to understand the effects of the individual variables on the loan support limit. Furthermore, the major variables with a high correlation with the loan support limit were selected through Pearson correlation analysis, and multiple linear regression analysis was performed. RESULTS: Simple regression models and multiple regression models were derived to predict the tractor loan support limit. The coefficient of determination and the root mean square error were calculated to determine the accuracy of each regression model. In the simple linear regression analysis, the coefficient of determination of the engine-power-based regression model was the highest (0.87), followed by weight, engine displacement, width, length, and height. Similarly, the root mean square error was the smallest in the engine-power-based regression model at 3,770,370 KRW. As a result of performing multiple linear regression analysis using engine power and weight, which exhibited a correlation coefficient of 0.8 or higher in Pearson correlation analysis, the coefficient of determination and the root mean square error were 0.88 and 3,699,940 KRW, respectively. CONCLUSION: As the multiple regression model with engine power and weight as variables has a high coefficient of determination and small root mean square error, it is considered the most suitable for predicting the tractor loan support limit. The developed prediction model can save time and greatly help the decision-making process of farmers for purchasing agricultural tractors.