Making a financial model more predictive than extrapolative is conceptually simple – model more the causes than the results. However, it is much more difficult than to conceptualize, in fact far outside the trained skill set of most financial analysts without a powerful tool, to simulate the actions of a company that produce the financial results.
The difficulty starts with the requirement that to be realistic, one cannot simulate each financial variable in isolation to each other. Consider revenue and the action of selling. If the action of selling generates revenue, it must decrease inventory and increase cash and possibly receivable in a specific way. Cash, inventory, receivables and revenue have to be modeled together. In fact, all the financial variables have to be connected to one another based on certain relationships. >>> The current tools that financial analysts utilize, such as Excel or similar free-form productivity tools, provide the ease to model either one or a small number of variables in isolation, but they are not the effective tools to model all the financial variables collectively.
The key to model these variables collectively is the realization that the way these variables are connected is such that as a whole they make up a “living” entity, so to speak, and when they change, as a result of an action by the entity, the variables do so only in ways that the entity retains its identity or “life”. So, instead of focusing on the variables and adding more and more formula to connect them, one models the host object that embodies all the variables as its attributes, and specifies the actions the host can perform. >>> In the case of our interest, the host object is a company, whose financial positions make up its attributes, and who performs only a set of specific actions in its business life. The software that enables analysts to model a company in this way will in my view be the next generation financial modeling tools.
Unnatural as it may be to a financial analyst, this way of viewing complexity is nothing new over the past fifty or more years to anyone with a beginner’s knowledge of object-oriented programming. So why has it not been implemented in financial forecasting? (Admittedly, it may well has been, but only confidentially to a small group.) My theory is that it is difficult to garner enough experiences in both fields in order to understand both in their utter most basics. >>> Imagine a programmer who wants to write a program for shape recognition. The biggest initial challenge is how to define as executable codes the concepts, such as a shape, so intuitive and self-explanatory that they appear to be undefinable. One has to have a great deal of life experience with shapes and shape recognition before he can extract the most essential features into executable attributes and actions.
I believe that the tool we provide in Modtris is taking the right step in this direction. Instead of modelling the financial variables disparately, Modtris models a company as an entity capable of certain basic actions. >>> To a financial analyst, Modtris is similar to an animator’s software tool that has ready-to-customize human figures and pre-programmed basic movements so that the animator needs only, as opposed to draw line by line and shade by shade, to choose and combine the basic elements to create complicated life-like animations. >>> Analysts using Modtris will lose some of the freedom he normally enjoys with the free-form tools, but gains the power to model a company in the way he understands in the real-world and turn his understanding into powerful long-term predictions.
To experience this way of financial modeling and forecasting, contact us at http://modtris.com, or write us at firstname.lastname@example.org. Modtris developer will work for free with you to build a Modtris model for a company you are covering in your research. You can see for yourself if Modtris provides abilities you haven’t had before.