Both analytical and mathematical (or inferential) tools are very useful evaluating accounting discrepancies and collecting evidence. In many ways they complement each other.
Analytical tools are very useful helping to chart and illustrate the process of an investigation. The use of flowcharts helps comprehend the flow of events sequentially. Many hard to grasp, abstract concepts can be concretized using visual aids.
Chart creation also enables easy collection of inputs from firm “informers”: insider people who can be a valuable input source. This helps formulate an accurate idea of what the firm is “up to.”
Whereas the inferential tools use mathematical relations and logical reasoning to arrive at conclusions. During an investigation, an inference model allows us to take into account all our variables at one: all of our evidence. Bayesian networks and Markov networks are two representations of inference networks. The networks link nodes, representing events that occurred in a causal relationship, using edges, forming a logical sequence.
A good example for a potential applicability of inferential tools was with the Bernard Madoff Ponzi Scheme. Enough evidence appeared every time fraudulent statements were back calculated and sent out to investors. A comparison between the share prices reported for the day in the statements mailed and the actual highs and lows could have caught early and prevented the crisis that unfolded.
