The article explores a type of model that allows for flexible dynamics and function form in analyzing financial data. The researchers developed a method using kernel smoothing and profiled likelihood to estimate the model. They found evidence of asymmetric news impact functions in their analysis of S&P500 index returns.