Towards a non-numerical component on decision-making risk management.


One of the most important aspects in finance is dealing with the uncertainty that all investments are subjected to. Asset managers and consultants try to achieve this seemingly difficult task by collecting raw numerical data, putting it through a descriptive and inferential statistical analysis in order to minimize risk incurred.

This whole process seems very objective and sophisticated, although this might not be the case. Consultants deal with corporations’ financial statements and try to assess their profitability, market over or under valuation, liquidity, internal and external debt exposure, etc. through the analysis of well-defined ratios which should provide all the meaningful and useful information that is out there.

On the other hand, asset managers deal with the investment portfolio construction. Once the filter of investable corporations is done, money managers’ work is to consider the risk that some investor is willing to take in his or her portfolio and maximize its expected returns. This process basically consists of statistically measuring how all these assets would move together (in fixed and/or changing proportions) using the portfolio standard deviation, which depends on the way the different assets are correlated, i.e. their mutually exclusive covariance. Consequently, volatility is supposed to be minimized.

Even though all the raw data used here is quantitative data, there is no empirical reason to strongly state that an active-managed fund will do better than a passive-managed fund (one that only follows some specific market’s movements) and this might be due to the fact that so very few qualitative variables are taken into account, although they still have a crucial role on our behavior, and therefore should be more responsibly weighted in our portfolio construction.

The main reason qualitative variables and human behavior should be reckoned with the risk management, is to outstand cognition biases and well psychologically-defined attention anomalies to which humans are exposed to, and therefore truly minimize our risk.

Cognitive biases

The main objective of this article is to allow the public to recognize and identify these attention anomalies that I will enumerate (Source: Yale University on Behavioral finance) :

Representativeness heuristic: It represents our tendency to see patters in what is really just a random walk. It is used when we judge the probability that an object or event A belongs to class B by looking at the degree to which A resembles B. When we do this, we neglect information about the general probability of B occurring (Kahneman & Tversky, 1972). Representativeness-based evaluations are a common cognitive shortcut across contexts.

Cognitive dissonance: It is the mental conflict that occurs when one needs to recognize that one’s beliefs are wrong. People tend to seek consistency in their attitudes and perceptions, so this conflict causes feelings of discomfort. Once you feel identified with a decision you have made, you just have a lot of trouble recognizing that it might not be an accurate one.

Wishful thinking bias: People is likely to exaggerate the probability of things that they feel identified with or that they desire to happen. They are often influenced by random successes and tend to attribute them to themselves.

Disjunction effect: It occurs when you need to decide whether you should or should not proceed with a course of action A that is somehow linked to a future event which has several outcomes. This uncertainty forces you to lengthen your time of response.

As an example, let us set two different outcomes here, TRUE or FALSE. You must decide whether to continue with the action A or do nothing. If you project yourself into the situation where outcome of the event is TRUE, then you prefer to continue with the action A. Now, if you project yourself in the situation where the outcome of the event is FALSE, you also prefer to continue with action A. You should therefore, rationally, continue with action A immediately. But in some context, a disjunction effect occurs and until you know whether the outcome of the event is TRUE or FALSE, you do not decide to continue with action A. Uncertainty causes you to delay your decision, even though you will continue to implement the same action once you know exactly the outcome of the event with certainty, whatever it may be.

Anchoring: This is a cognitive bias that occurs when people over-rely on an initial piece of information to make subsequent judgments in a decision-making process. Once a particular anchor point is established, people will unknowingly base their arguments and opinions on that point and only on that point.


To sum up, all these qualitative variables could be added into our calculations by their likelihood and then be weighted by the use of indicator variables, accordingly we would take into consideration some non-numerical data and transform them into a coherent mathematical language, minimizing future investment uncertainty, which is the core purpose of risk management.



Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.