Quantitative analysis of start-up companies: statistical and machine learning perspectives

Projet scientifique

The main objectives are:

• Provide methodological expertise for building a set of variables that could be used as covariates
• Provide expertise and guidance for choosing the type of model and/or an algorithm to link the covariates (also called predictors or explanatory variables) to the outcomes of interest
• Provide expertise and guidance for finding procedures that allow the models and/or the algorithms to select and combine in an ‘optimal’ way the `significant’ covariates in order to build accurate predictions. The financial and non-financial information collected on the start-up companies and transformed in covariates lead in general to a large set of such explanatory variables
• Propose metrics to assess the performance of the modeling and to decide when updates of the models and/or algorithms are needed