Abstract

Marketing mix modelling (MMM) is an important technique in marketing measurement to define effectiveness and efficiency of marketing investment. MMM is a complex combination of various statistical models and data transformation functions. Traditional MMM is subject to analyst bias and is slow and expensive. This disclosure describes automated techniques that reduce the analyst bias in MMM and make the model building process fast, flexible, and inexpensive. The techniques described in this disclosure overcome analyst bias through a multi-objective hyperparameter optimization that is achieved by using an evolutionary algorithm.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Share

COinS