Managers in 3G wireless markets require sales response models that provide substantive insights into the effects of marketing activities as well as reliable sales forecasts. In such markets the effects of service attributes and marketing activities could rapidly change over time; moreover, buying rates at individual consumer level often exhibit non stationarity. In recent years, many marketing scholars recognized that forecasting models based on bernoullian data seem to be a good tool in order to capture the purchasing process properties mentioned above. However, many companies collect cumulated (rather than bernoullian) purchasing data and they are unable to directly apply such high forecasting performance models. In this paper we first develop a stochastic Data Distribution Model (DDM) that transforms an original dataset of cumulated purchasing data into bernoullian infra-period observations and then apply a Dynamic Changepoint Model (DCM; Fader et al.2004) that captures the underlying evolution of the buying behaviour associated with the new mobile service. The DDM is based on the marketing principle that as consumers gain more experience with the service, we would expect their preferences (and therefore their sequence of purchases) to stabilize to some extent. We employ data on a consumer panel in which we track the monthly purchasing of each panelist for the duration of the test market and redistribute the cumulative data within each day according to the stochastic model. We then apply the DCM to a number of versions of the modified dataset. Parameters are obtained as the mean of simulated parameters for each application. We find that the model applied to an highly innovative service show a considerably strong empirical performance.
An Attitude-based theoretical model to explore the consumer adoption of 3G Mobile TV Services
PAGANI, MARGHERITA
2006
Abstract
Managers in 3G wireless markets require sales response models that provide substantive insights into the effects of marketing activities as well as reliable sales forecasts. In such markets the effects of service attributes and marketing activities could rapidly change over time; moreover, buying rates at individual consumer level often exhibit non stationarity. In recent years, many marketing scholars recognized that forecasting models based on bernoullian data seem to be a good tool in order to capture the purchasing process properties mentioned above. However, many companies collect cumulated (rather than bernoullian) purchasing data and they are unable to directly apply such high forecasting performance models. In this paper we first develop a stochastic Data Distribution Model (DDM) that transforms an original dataset of cumulated purchasing data into bernoullian infra-period observations and then apply a Dynamic Changepoint Model (DCM; Fader et al.2004) that captures the underlying evolution of the buying behaviour associated with the new mobile service. The DDM is based on the marketing principle that as consumers gain more experience with the service, we would expect their preferences (and therefore their sequence of purchases) to stabilize to some extent. We employ data on a consumer panel in which we track the monthly purchasing of each panelist for the duration of the test market and redistribute the cumulative data within each day according to the stochastic model. We then apply the DCM to a number of versions of the modified dataset. Parameters are obtained as the mean of simulated parameters for each application. We find that the model applied to an highly innovative service show a considerably strong empirical performance.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
