Understanding consumer behaviour
We have developed important statistical and mathematical models for forecasting consumer buying behaviour.
Predicting consumer buying behaviour is key to the activities of manufacturing and retail organisations. Mathematical modelling is an important element of this process. It helps companies make important decisions about their products.
Researchers from our School of Mathematics worked with Nielsen to develop statistical and mathematical models for forecasting consumer behaviour.
By studying the statistical properties of classical models we were able to confirm that the theories being applied by Nielsen were a viable foundation for effective forecasting. This research also provided valuable insights into the most efficient methods for analysis and estimation.
Testing statistical models
The statistical model in question is the so-called Polya process.
The potential of this model for use in market research had been debated for many years. However, discussions never progressed beyond a basic level and as a consequence the model had never been tested and used on data sets of practical value.
Our researchers extensively studied the statistical properties of this model with particular emphasis on market research applications.
In the course of developing the statistical methodology we tested the methods and models on the largest set of household panel data ever analysed, courtesy of Nielsen.
Nielsen is an American global information and measurement company. It operates in Africa, Asia, Australia, Europe, Middle East, North America and South America. From 2011-2014, the company has consistently ranked first in the Honomichl Top 25 largest market research organisations in the world.
Applying models to data
Nielsen has used the Cardiff research to provide services to a host of major corporations. These include Unilever, Coca Cola, Pepsi, Kraft and Nestlé.
The research has been key to the work conducted by Nielsen for these global organisations and has enabled the consolidation and progression of corporate relationships. Subsequently, this has enabled Nielsen to compete on a worldwide scale and retain its dominant market position.
Related research by Professor Zhigljavsky and his team has had significant impact for other companies, such as Procter & Gamble. The models developed at Cardiff University were applied to study sales data versus pricing in several European countries. Our models provided more reliable and accurate forecasts compared with those previously used.
The research has been presented by Professor Zhigljavsky at over twenty national and international events, enhancing the knowledge and ability of leading organisations in the manufacturing and retail industries to implement more effective forecasting strategies and respond to a rapidly changing economic climate.
- Zhigljavsky, A. A. 2011. Statistical Modelling in Market Research. In: Lovric, M. ed. International Encyclopedia of Statistical Science. Springer Reference Berlin: Springer, pp.1450-1452. (10.1007/978-3-642-04898-2_548)
- Savani, V. and Zhigljavsky, A. A. 2007. Asymptotic distributions of statistics and parameter estimates for mixed Poisson processes. Journal of Statistical Planning and Inference 137 (12), pp.3990-4002. (10.1016/j.jspi.2007.04.016)
- Leonenko, N. N. , Savani, V. and Zhigljavsky, A. A. 2007. Autoregressive negative binomial processes. Annales de l'Institut de Statistique de l'Universite de Paris 51 (1), pp.25-47.
- Savani, V. and Zhigljavsky, A. A. 2006. Efficient parameter estimation for independent and INAR(1) negative binomial samples. Metrika 65 (2), pp.207-225. (10.1007/s00184-006-0071-x)
- Savani, V. and Zhigljavsky, A. A. 2006. Efficient Estimation of Parameters of the Negative Binomial Distribution. Communications in Statistics - Theory and Methods 35 (5), pp.767-783. (10.1080/03610920500501346)