의사결정나무 분석을 이용한 국내 프로야구선수들의 연봉 결정요인 분석

The purpose of this study is to present a predictive model of the determinants for Korean professional baseball players’ salaries. 265 Players registered in 2015 Korea Professional Baseball Organization (KBO) were selected as research data and Decision tree analysis using CART algorithm was conducted. The variables used in this study were separated from the last year (2014) and career performance, pitcher and batter. Therefore, 4 predictive models were analyzed and the results of each model are as follows. First, determinants of batters’ salary in accordance with the last year (2014) performance were batting average, base hit, and career. Determinants in accordance with the career performance were the number of game play, base hit, and batting average. Second, determinants of pitchers’ salary according to the last year (2014) performance were career, ERA, and strikeout. Also, determinants according to the career performance were ERA, win, and the number of game play. Compared with the results of previous studies, point (difference or specificity) noted in this study are major players batting over 0.257. They were re-separated from base hit 173.5 and long-distance hits can be interpreted as a key variable determining high salary among the high average hitter. The ERA was the same in determinants of pitchers’ salary according to the last year( 2014) and career performance. But, they were separated from different criteria. It also means that the evaluation standard for the performance of the players in the salary negotiation process varies depending on the last year and career performance. It suggests the need for a salary negotiation considering both sides.