What's Hot

Nassim Nicholas Taleb's blog, an inspiring read | Incerto

Monday, April 9, 2018

Applied Markov Example

Dear Causal Capital, we have been reading the ISO 31010 guideline on a Markov Matrix in section B.24 of the risk assessment techniques guideline and the concept is a touch abstract to understand. Do you have an example on where this risk analysis technique can be used to improve decision making in the business, something that goes beyond just Risk Assessment from a Risk Management perspective.

Okay, in this short blog we'll take a quick look.

What is a Markov Matrix
Perhaps we should begin with what a Markov Matrix is not and with that, it must be stated that a Markov Matrix has nothing to do with the infamous Risk Matrix that is used to summarise threats in a quadrant report that is presented to executives at disparate intervals over a typical year. On the other hand, the Markov Matrix can be seen as a two-factor analysis table that allows stakeholders to compare various 'assets in a portfolio' and anticipate what their future state might be, given the prevailing net uncertainty of the portfolio.

A Markov HR Example
In the diagram below you can see a sample spreadsheet showing employees (our matrix of assets) residing in various job functions across a hypothetical enterprise. Each year specific dynamics play out (our risk needing assessment) that have the potential to leave one department under resourced while other business functions overwhelmed with staff. These dynamics of uncertainty may include but are not limited to: retiring employees, departmental growth requirements, job leavers, job hoppers, staff being fired and new entrants.

So how to model this combined net level of human resource uncertainty and see the big picture in it all? Well, the Markov Matrix does this for us and at an enterprise-wide level, and in this case, it will allow a human resource function to identify how many and the types of resources they need to decide to have in their plan so that the company can reach its objectives.

The Markov Matrix could be applied to a huge array of different business risk modelling requirements such as procurement or operations planning, even sales coverage could potentially benefit from this top level but detailed analysis of 'asset migration' in an out of the working space.

Example Markov Analysis | Causal Capital

The benefits of the Markov Matrix are numerous when one thinks about it, starting out with it's very easy for risk managers to quickly assemble one of these models. Also the Markov Matrix supports the integration of a Monte Carlo simulator without violating the structure under the matrix. Finally, if you were to study and account for the correlation or co-variance of specific asset positions in relation to the total portfolio, the Markov Matrix can be adapted to take in the sum product of migrations across the matrix through time.

Anyway, we are a big fan of the technique in Causal Capital and I might see if a working spreadsheet can be pulled together, perhaps one with the Monte Carlo plug-in and correlation system turned on. 

For those attendees that are going to be part of the QRA Masterclass coming up in the UAE, this risk modelling technique will be discussed in detail.

No comments:

Post a Comment