The unbroken circle:
MetadataShow full item record
In this dissertation I construct a model of corporate political activity, which I use to explore the relationship of corporate political activity related to taxation and corporate tax aggressiveness. In doing so, I integrate literature from the disciplines of political science and accounting. I hypothesize that firms optimize their tax planning and political strategies jointly, and that this decision-making process follows a specific dynamic pattern. In order to test my hypotheses, I first estimate latent variable models representing the constructs that I develop. I then estimate a series of models using a variety of estimation techniques, including Ordinary Least Squares (OLS), the Generalized Linear Model (GLM), Kernel Regularized Least Squares (KRLS), LASSO regression, and Random Forests. I evaluate the validity of the latent variable estimations and the extent to which the estimations support the hypotheses that I generated.