Applying GRA and path analysis jointly to analyzing capital structure

Both capital structure and its antecedent variables are unobservable constructs measured by a set of observable indicators or proxies. Traditionally, the ordinary square (OLS) method was often used for the research on capital structure. However, the use of the ordinary squares method faces some problem. For overcoming these problems mentioned above, we use Grey relational analysis and path analysis jointly to analyze capital structure. In this study GRA is used to obtain the integrated indexes of three latent variables, i.e. capital structure, growth and profitability, and path analysis is used as a method for studying the direct and indirect effects of variables. Consequently, we verified the negative relationship between profitability and leverage, the negative relationship between growth and leverage, and the mediating role of profitability between growth and leverage. In addition, our results reveal that growth is a more important determinant for capital structure than profitability. We developed a new possible hybrid method to test the relationship between capital structure and this method is more convincing and powerful.