This paper investigates the effect of aggregation and non-linearity in relation to television rating forecasts. Several linear models for aggregated and disaggregated television viewing have appeared in the literature. The current analysis extends this work using an empirical approach. We...

We compare three methods of estimating the duration, or half-life, of how well each method works with the data aggregated over different time intervals. In contrast with the existing theory on the, so called, data interval bias, our experiments are based upon realistic advertising schedules. Our...

This paper is aimed to provide an economic interpretation on the emergence and evolution of the specialised middlemen whose duty is to facilitate the transactions of goods and services in an economy. In a general equilibrium framework, the emergence and evolution of the specialised middlemen...

This article characterizes the conditions under which holdout (i.e. bargaining inefficiency) may, or may not be significant in a two-sided, one-buyer-many-seller model with complementarity. Our central result is that the severity of holdout (i.e. inefficiency) is critically dependent on three...

We describe some fast algorithms for reconciling large collections of time series forecasts with aggregation constraints. The constraints arise due to the need for forecasts of collections of time series with hierarchical or grouped structures to add up in the same manner as the observed time...

Multi-step forecasts can be produced recursively by iterating a one-step model, or directly using a specific model for each horizon. Choosing between these two strategies is not an easy task since it involves a trade-off between bias and estimation variance over the forecast horizon. Using a...

Theoretical results on the properties of forecasts obtained using singular spectrum analysis are presented in this paper. The mean squared forecast error is derived under broad regularity conditions, and it is shown that the forecasts obtained in practice will converge to their population...

One of the most widely used standard procedures for model evaluation in classification and regression is K-fold cross-validation (CV). However, when it comes to time series forecasting, because of the inherent serial correlation and potential non-stationarity of the data, its application is not...

This paper investigates the empirical properties of autoregressive approximations to two classes of process for which the usual regularity conditions do not apply; namely the non-invertible and fractionally integrated processes considered in Poskitt (2006). In that paper the theoretical...

We show how cubic smoothing splines fitted to univariate time series data can be used to obtain local linear forecasts. Our approach is based on a stochastic state space model which allows the use of a likelihood approach for estimating the smoothing parameter, and which enables easy...