<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=1406241429584765&amp;ev=PageView&amp;noscript=1">


Forecasting Earnings Surprises with Machine Learning

The goal of this study is to see if the Earnings Per Share (EPS) estimates provided by IBES can be used to forecast the actual EPS of a company with Machine Learning. However, since EPS announcements are quarterly, each company will only have a couple of dozen values which is insufficient for training a Machine Learning model that predicts a specific company’s EPS.

Instead of predicting whether a company will beat or miss an estimate we can instead predict whether the analysts’ estimate will be right or wrong. Since many companies have IBES estimate values we will have enough data to train a model. This does mean that we will have to make some transformations to the IBES data so that the models are company agnostic.

Download our guide to uncover how the spreads between analyst estimates impact their EPS forecasts. 
IBES Study book

Download Now