Social and political relationships are generally characterized by nonlinearity and complexity, and are usually of unknown functional forms. Many statistical methods employed in social science research, such as multi-l...Social and political relationships are generally characterized by nonlinearity and complexity, and are usually of unknown functional forms. Many statistical methods employed in social science research, such as multi-linear regression, logistic regression, logit and probit models, always assume exact and, in general, linear functional forms for the underlying utility functions. Neural network models, on the other hand, are capable of approximating arbitrary functional forms under general conditions and can handle rich patterns of nonlinearity in the utility functions. This article suggests that neural network models can be widely applied to social science statistics.展开更多
文摘Social and political relationships are generally characterized by nonlinearity and complexity, and are usually of unknown functional forms. Many statistical methods employed in social science research, such as multi-linear regression, logistic regression, logit and probit models, always assume exact and, in general, linear functional forms for the underlying utility functions. Neural network models, on the other hand, are capable of approximating arbitrary functional forms under general conditions and can handle rich patterns of nonlinearity in the utility functions. This article suggests that neural network models can be widely applied to social science statistics.
