Predicting IPO initial returns using random forest

摘要: Abstract Empirical analyses of IPO initial returns are heavily dependent on linear regression models. However, these models can be inefficient due to its sensitivity outliers which common in data. In this study, the machine learning method random forest is introduced deal with issues cannot solve. The used predict IPOs issued Borsa Istanbul. prediction accuracy then tested against methods robust regression. results show that has by far outperformed other every category comparison. variable importance measure shows proceeds and volume most important predictors returns. also variables act as potential proxies for ex-ante uncertainty more than information asymmetry.