both conceptual and mathematical bases for spatial prediction, such as spatial interpolation, spatial gravity modeling, spatial regression, and spatial optimization. These traditional analog and mathematical modeling techniques are commonly taught in geography, geology, epidemiology, criminology, civil engineering, transportation science, urban and regional planning, and landscape architecture departments. A few universities offer advanced geocomputational methods for spatial prediction, such as Monte Carlo simulation, Markov chain modeling, cellular automata, agent-based modeling, geographically weighted regression, spatial self-organizing maps, spatial trajectory modeling, spatial niche modeling, spatial Bayesian statistics, and spatial econometrics. Example universities offering courses in the spatial aspects of forecasting include Arizona State University; Clark University; the University of Texas, Dallas; San Diego State University; the University of Utah; the University of Maryland; and Ohio State University.
Some community colleges or technology centers (e.g., GeoTech Center) offer basic statistics courses or computer modeling tools (such as STELLA), which can provide foundation training for beginners. Opportunities for professional training in forecasting are limited. Workshops or summer schools, such as those offered by the Spatial Perspective to Advance Curricular Education program,10 the Center for Spatially Integrated Social Science,11 and the University of Michigan, are perhaps the main form of training for advanced space-time methods or geocomputational techniques. Many of these workshops cover only the fundamentals. For economics and business, the IIF frequently offers training workshops for practitioners at their conferences.