A variable impact neural network analysis of dividend policies and share prices of transportation and related companies

Abstract

The purpose of this research is to investigate dividend policy, including its impact on share prices of transportation providers and related service companies, by comparing generalized regression neural networks with conventional regressions. Our results using regressions reveal that for Europe and for the US and Canada the market-to-book-value, as a surrogate for growth opportunities, fulfils expectations of pressures on dividends leading to a negative association with dividend yields in accordance with the pecking order theory. Neural network analysis indicates a clear role for growth opportunities for the US and Canada pointing to an underlying confidence on the part of transportation companies in their own internal policies. Finally, risk is rewarded especially in Europe.