Does learning by disaggregating accelerate learning by doing? The effect of forecast disaggregation on the year-over-year improvement in demand forecast revisions.


Speaker


Abstract

Demand forecast accuracy is critical to organizational planning and coordination, and forecast accuracy is an important source of competitive advantage. Indeed, surveys indicate that CFOs name forecast error as their top internal concern and identify demand forecasting as one of their top organizational priorities. Moreover, a competitive advantage accrues to organizations that, not only forecast more accurately, but also earlier in the forecast period. In this study, we examine whether the disaggregation of a demand forecast into separate forecasts for each source of demand accelerates the improvement in forecast accuracy over time in managers’ forecast revisions. We first document that forecast revisions improve through learning by doing: The decline in forecast error with shortening forecast horizon (e.g., from 9-month to 1-month) increases with experience. We then hypothesize that disaggregation of the forecast accelerates learning by doing. Exploiting our unique access to proprietary data from a multinational manufacturing organization we find evidence that is consistent with our predictions. Our study provides evidence of how a change in the way in which managers formulate and communicate forecasts – that is, forecasting different sources of demand separately - can help address the vexing problem of demand forecast error.

Zoom link: https://eur-nl.zoom.us/j/98094168133?pwd=TE56WWRZdzcrNkFUVTNXV1hEeW92QT09