Predictive Science

Author: R. Scheper (France) et al.
Translator: Chen Yanping
Editor-in-Chief: Wang Hongliang
Publisher:
Publish Date: 2005-03-01
Features:
Excerpt: You prefer approximate accuracy over precise errors when designing plots, don’t you? Without a doubt, I lean toward the former. A plot actually consists of three points: foundation, progression, and the final scenario. The foundation refers to a representative of a realistic situation that maintains vitality in the long term (not referring to annual budgets or election deadlines). Then, as we move forward in time, we gradually build a possible future tree. For example, Europe is expected to achieve monetary unification around 1997 (or at least by 1999). So, can we be certain about when this currency will be introduced? If not, we must resort to "what-if" reasoning: If the five countries (Belgium, the Netherlands, Luxembourg, Germany, France) achieve monetary unification—if we start in 1997 or 1999—if we decide not to unify currencies at all—if we unify in 2000—each assumption leads to a different plot development. Therefore, the key lies in expanding the realm of possibilities. But won’t we get lost in multiple conflicting scenarios? What is the standard for trusting a scenario? Why do we prefer to believe one scenario over another? You raise a valid point: the complexity of the path often leads us to imagine many scenarios, which not only fails to help decision-makers but also leaves them overwhelmed. This necessitates reduction, leaving only a few limited, yet fully indicative, scenarios for possible futures. So, how do we filter them? There’s controversy here too. Some say, "We should calculate the probabilities" and develop mathematical tools for this purpose (though these tools only process information derived from initial expert opinions). Others, including myself, believe that mathematical operations on probabilities play a minimal role—the key lies in clearly articulating the choices you make. If you want your predictive scenarios to be useful to a rational decision-maker, you should introduce three to five possibilities, but absolutely not a thousand—otherwise, the sheer volume of information from today’s media will leave them overwhelmed.