Predicting Post-merger Performance of US Banking Industry

This paper investigates the ability of neural network models to predict the post-merger performance of mergers and acquisitions (M&As) in US banking industry. As we known, this is probably the first empirical study applying neural networks in this topic. The aim is to offer an alternative tool for making M&A decision from the view of potential synergy effect and improve the rate of success on M&As deals. This study first provides a detailed discuss from synergy effect and strategic fit. It then develops and compares the forecasting performance of regression and neural network models. The results show that the ability of neural network models to catch nonlinear relationships and complex interactions between amounts of data and factors is potentially fruitful for evaluating M&As synergy effect. However, neural networks have been criticised as not human understandable for being a black box. To solve this problem, sensitivity analysis is used to explore the relationship between independent variables and dependent variables