Forecasting risk of decision – making processes

What is the risk of decision-making processes, what causes it? Most typically, definitions of risk are ex post – they are looking at risk as a difference between expectations of results of actions taken and the actual performance. This is a considerable inconvenience, especially in cases when processes are of a long-term nature. Thus, is it possible to measure risk in the course of the decision-making processes? How can this be done and in what conditions can risk measures be extrapolated? An analysis of the definitions of risk shows that the one which is the most useful for solving the problem undertaken in the present study, is given by K. and T. Jajuga, “…the term of risk will refer to a decision, or, to be more precise, to an action taken as a result. One may therefore speak of taking risky decisions”. The authors make it clear that a risky decision is uncertainty as to “…the possibility for people to control the factors that determine the reality”. This suggestion, if accepted, enables one to construct a risk model as a random vector whose components are control variables of the decisionmaking processes taking place. In consequence, this makes it possible to estimate statistic measures of risk. Risk measures indicating the level of risk at moment t of decision-making processes represent a foundation of the problem announced by the title of the present study. Although they are merely a set of risk estimations, i.e. an assessment of its state, they nevertheless provide an opportunity to forecast risk levels within the period in which the processes occur, thus providing valuable information for decisions-makers.