Multivariable panel data ordinal clustering and its application in competitive strategy identification of appliance-wiring listed companies

To fully extract information of panel data from special dimension and time dimension, we provide a multivariable panel data ordinal clustering method based on the Fisher ordinal cluster theory. The innovation of this paper is that we reconstruct Ward function by Frobenius norm and make empirical research on identifications of competitive strategies and strategic evolution by using the financial data of appliance-wiring listed companies. We first reduce strategic variables of total asset turnover ratio, fixed asset turnover ratio, Gross margin, R & D propensity and turnover ratio of operating expense on income by factor analysis and make ordinal clustering according to factor scores. Then, we find the turning point in the fitted curves of pure technical efficiency calculated by Malmquist DEA for the companies that are consistent with the strategic turning points of the ordinal clustering result. The empirical results illustrate that the method is valid in identifying competitive strategies and identifying the time at which competitive strategy began to evolve. In sum, both low-cost and differentiation strategy exist in the appliance-wiring industry, but, some listed companies have no obvious strategy. Strategic evolution exists in some appliance-wiring listed companies.