A nearest neighbor model for forecasting market response

摘要: Abstract Researchers in marketing often are interested modeling time series and causal relationships simultaneously. The prevailing approach to doing so is a transfer function model that combines Box-Jenkins with regression analysis. component assumes stationary, stochastic process generates each data point the series. We introduce multivariate methodology uses nearest neighbor technique represent behavior complex nonstationary. This represents deterministic as discrete dynamic system. In this paper we describe how may exhibit chaotic behavior, present method capable of representing such behavior. provide an empirical demonstration using store scanner for consumer packaged good.