Dynamic Bayes Modeling

Home » Software: Marketscience Studio » Dynamic Bayes Modeling

The core modeling module provides a comprehensive user interface (UI) for specifying a range of dynamic linear panel models at all levels of the client business. User-specified model structures are integrated with the model database to perform any required variable transformation which is then transferred to the proprietary model estimation algorithm housed within the Oxmetrics analytics package. Results from Ox are then returned for viewing in the UI with full graphical interface

Underlying the power of the Marketscience Studio is the unique estimation algorithm developed using the State Space Functions Pack (SSF Pack) written and supported by Professor Siem Jan Koopman. Marketscience have partnered exclusively with Professor Koopman to accommodate the unique needs of modeling consumer demand as an evolution of the traditional Marketing Mix Modeling approach.

SSF pack is the original and leading set of algorithms for estimating the parameters of a wide array of time series models based on the Kalman filter recursions. Our development now adds the Bayesian Pooling capability to enable more rigorous general to specific modeling within large panel data sets. This provides a highly flexible modeling framework, incorporating baseline and marketing dynamics together with a rich specification of consumer response to marketing investments: from full heterogeneity, through common geographical regions and/or product levels up to fully pooled as required.

  • 1

    Complete heterogeneity of response across all levels of the model hierarchy

  • 2

    Full or partial shrinkage depending on response patterns across defined segments

  • 3

    Resulting in the most parsimonious model possible whilst still maximizing insights

Model results viewing (in ox)….

In addition to the estimation approach our UI incorporates all of the major data transformations necessary to cover all potential marketing response functions

  • Polynomial Distributed Lag (PDL) structures

  • Adstocks

  • Diminishing returns

Data Visualization & Analysis

Dynamic Bayes Modeling

Full System Decomposition

Simulation & Optimization

Forecasting & Reporting

Scroll to Top