Distributed investment decisions and forecasting errors: An analysis based on a multi-agent simulation model

This paper analyzes the impact of both non-systematic and systematic forecasting errors on the coordination of distributed investment decisions. The forecasts of concern pertain to the expected cash outlay necessary to launch and operate an investment project, to the expected cash flows generated by the project's operation, and to the self-assessment of whether or not the abilities expected to be needed for operation coincide with one's own. Systematic forecasting errors manifested by systematically over- or underestimating these predicted project indicators are interpreted as being produced by overconfidence. Utilizing a multi-agent simulation approach, we show that within a hierarchical business organization with distributed authorities for decision making, in some cases being overconfident can mitigate the negative effect of erroneous forecasting compared to being non-systematically wrong.