Skip to main content

Participation in external events Machine learning versus conventional static methods

Are machine learning methods more suitable than conventional statistical methods to predict IPO withdrawal? Dr Annika Reiff rxplored this question in her presentation "Comparing the prediction performance of random forest, lasso and logit in the context of IPO withdrawal" at the World Conference on Data Science and Statistics, which took place in mid-June in Amsterdam.

The results of the analysis show that under certain conditions more complex machine learning methods are indeed better suitable to predict whether a firm will withdraw its IPO.