Merger Simulation with Brand-Level Margin Data: Extending PCAIDS with Nests

We present a method to calibrate empirically the demand parameters in a merger simulation model by using brand-level profit margin data. While the approach can be generalized, we develop these ideas within a particular framework - the PCAIDS (proportionality-calibrated AIDS) model. We show that the brand-level margins effectively define product "nests" (products that are especially close substitutes) and substantially increase the flexibility of PCAIDS for modeling critical own- and cross-price elasticities. The model is particularly valuable for transactions at the wholesale level (where scanner data do not exist) and for geographic markets that span national borders (where comparable data may not be available), since other methods to derive elasticities, particularly those based on econometric estimation, may not be possible or may not be reliable.