Forecast/forecasting aggregate

Revenue management is focused on demand forecasting - aggregated and disaggregated - demand distribution models or arrival processes to... 

Alon, I., Qi, M. and Sadowski, R. J., 2001. Forecasting aggregate retail sales a comparison of artificial neural networks and traditional methods. Journal of Retailing and Consumer Services, 8(3), 147-156. 

As stated at the beginning of the chapter, the success of a manufacturing supply chain depends on its ability to accurately forecast customer demands and plan the production process to meet those demands in time. So far in this chapter, we discussed the forecasting principles and several qualitative and quantitative methods for demand forecasting. Based on the demand forecasts, aggregate production planning is done to plan the production. 

Vol. 284 H. Liitkepohl, Forecasting Aggregated Vector ARMA Processes. X, 323 pages. 1987. 

The other approach is the bottom-up procedure, which proceeds to aggregate projected sales data. You start with the projected sales data in each territory for each product and sum up the forecasts into successively larger amalgamations. 

Demand is forecasted with price and quantity in a first step and aggregated to a total demand volume with an average price. 

The rolling planning process starts with a first phase ( ) typically in the first half of the period. The regions update the regional demand forecast by price and quantity that is aggregated to a global demand. In analogy, offers... 

Elasticity analysis algorithm as pragmatic approach to determine average price elasticity of aggregated demand forecasts that analyzes underlying customer price-quantity demand. 

From our experience in the pharmaceutical world, however, these types of forecasting analyses boil down to fwo sefs of aggregate assumptions about an indication marketplace and the position of fhe NCE wifhin it, as shown in Figure 35.3. 

If the upward bias of aggregate forecasts over the period were estimated to be 5%, for example, the corresponding cutbacks relative to these forecasts would be about 25% and 10%, respectively. 

Unpredictable variability in individual responses, coupled with the need to forecast the aggregate responses of an entire population of future subjects, provide the reasons why biostatisticians are involved in clinical trial design and analysis. Inferential statistics is the discipline of making inferences about populations by analyzing data from samples that were drawn from those populations in a prescribed way. If we could somehow look into a crystal ball and measure the actual future responses to a new drug from the entire... 

Two methods [6-10] are used to obtain aggregate impacts from overlays (a) conventional weighting, the weights being a measure of relative importance (b) threshold technique, in which a unit square is excluded from further consideration whenever a designated number of impacts are forecast to occur, or whenever an individual impact is unacceptably high. 

Using daily exchange rates as opposed to weekly or monthly exchange rates ensures the convergence of GARCH parameters and minimizes standard errors. The aggregation formula for monthly GARCH forecasts is ... 

The facilitator. .. produces a summary of the panel s forecasts, along with their justifications for the forecasts. The panel members are anonymous in this summary (and any subsequent summaries.) The summary is then returned to the panel members, and they are allowed to update their forecasts (if they choose) and send a new forecast (along with justifications, counterarguments, etc.) to the facilitator. This process may be repeated for more rounds. Usually, the panels forecasts converge over the rounds. A compound decision rule (one rule for when to stop, and another for how to statistically aggregate the final set of forecasts) ends the process. ... 

Similar to the aggregation methods, communication levels between the participants in prediction markets tend to be near zero. Therefore, the prediction markets also tend to be less affected by inhibitory influences. It is possible that prediction markets may be vulnerable to herd behaviour whereby individual participants may follow the actions of the majority - the herd - lessening the degree of accuracy of the market s forecasts. This may also be further exacerbated by the overconfidence of individual participants. Yet such negative influences would also create opportunities... 

The discussions cover the five main techniques employed in forecasting, ranging from unstructured FTF meetiugs to structured NGT, from statistical aggregation to the Delphi method and to prediction markets. They are, by all means, not exhaustive h° The discussiou of these five techniques does not intend to champion one technique over the others, but iustead is intended to inform the reader of the possibilities with the hope that further discussion can ensue on improving this third branch of a regulatory regime. 

---