Empirical modelling is a set of techniques such as statistics, econometrics and data science more in general where we analyse mainly quantitative data to obtain precise and meaningful estimates of an issue. Well executed empirical modelling goes beyond a simple numeric description of a phenomenon and provides insights on the causal effects or at least the descriptive relationships among variables.
Empirical modelling can and should be used whenever data is available to analyse a phenomenon. Empirical modelling allows us to go beyond the simple computation of averages and proportions by estimating models which allow us to better understand the causal relationships among variables.
Empirical modelling can be used to quantify and monitor relationships and phenomena. In this sense, this approach can be used in almost any project both as a primary modelling approach and as a secondary approach to assist other modelling techniques such as the computational simulation models.
Good empirical modelling critically depends on two things: good and reliable data on the one hand and a good understanding of the analysis issue on the other hand.
CORESO assists partners in identifying the good data sources and helps prepare the data for a sound analysis.
To improve accuracy of our empirical models we closely collaborate with our partners and use collective intelligence techniques to gain a good understanding of the analysed phenomenon.