Most people think so and they revise forecasts from quantitative METHODS, usually reducing accuracy as a result. Nevertheless, people often have useful KNOWLEDGE about the problem, which is REFERRED to as domain knowledge. One approach to making effective use of domain knowledge consists of providing graphic decision support for judgmental forecasting (Edmundson 1990). Another approach is to integrate domain knowledge with statistical methods. For a review of RESEARCH in this area, see Sanders & Ritzman (2001).The best way to integrate judgment with statistical methods is as an input to the quantitative models For example, causal-force knowledge can be used to incorporate knowledge about trends into forecasts (Collopy and ARMSTRONG 1992, and Armstrong and Collopy 1993).

Most people think so and they revise forecasts from quantitative methods, usually reducing accuracy as a result. Nevertheless, people often have useful knowledge about the problem, which is referred to as domain knowledge. One approach to making effective use of domain knowledge consists of providing graphic decision support for judgmental forecasting (Edmundson 1990). Another approach is to integrate domain knowledge with statistical methods. For a review of research in this area, see Sanders & Ritzman (2001).The best way to integrate judgment with statistical methods is as an input to the quantitative models For example, causal-force knowledge can be used to incorporate knowledge about trends into forecasts (Collopy and Armstrong 1992, and Armstrong and Collopy 1993).