Identifying Expensive Trades by Monitoring the Limit Order Book

This paper provides an analysis of the impact of microstructure variables on the transaction costs for split orders on 171 Euronext large cap stocks. First, using the adaptive Lasso selection method, we conduct an exploratory study to identify which microstructure variables best explain the transaction costs of split orders. Then, we propose an ordinal logistic model to classify ex ante transaction costs into buckets. Our empirical work demonstrates that our model performs very well both in‐sample and out‐of‐sample. All the findings show that microstructure information is of great importance for any investor who attempts to manage the execution of split orders throughout the day.