The application of Kalman filtering to corporate bankruptcy prediction

In this paper we propose a state space model to predict failed companies based on Kalman filter theory which is creative in the field of finance. A Kalman filter is simply an optimal recursive data processing algorithm which is in the form of a set of equations that allows an estimate to be updated once a new observation becomes available. Given a set of parameters (mainly of financial nature) it describes the situation of a company over a given period, and predicts the probability that the company may become bankrupted during the following year. It is clear that probabilistic models are better suited for class distribution prediction. Also this type of program provides an output in the form of a decision with given functions and data. We can treat it like a computer program which returns an answer depending on the input, and more importantly, it can potentially be inspected, interpreted and re-used for different situations. The model fits the data well and gives a sensible answer to the actual bankruptcy prediction problem.