기업성장에 관한 분위수 접근방법

The goal of this paper is to suggest an appropriate method for the studying of firm growth rate in a Gibrat's Law context. Instead of using traditional OLS regression method that is only appropriate for normal distribution, this work uses the linear quantile regression method which does not assume any particular distributional form of the dependent variable and investigates all defined values of the dependent variables. The work focuses on two respects. First, we investigate the complete shape of the distribution of firm growth rate, rather than limiting the analysis about conditional mean growth level. Second, we show that the differences in the firm growth rate process between large and small firms are highly circumstantial. Also, industry dynamics have a substantial influence on the relationship between firm size and firm growth. The data used includes more than 509,217 financial data of Korean firms (Kis-value data). We provide significant evidence indicating that firm growth studies should look to other parts of the firm growth rate distribution instead be less obsessed with explaining means. Ultimately we can suggest some significant policy implications and improvements with this results.