Chapter 6 Quantifying Risk through Modeling

 

Quantifying Risk through Modeling

It is better to be approximately right than to be precisely wrong.

HOW NOT TO QUANTIFY RISK

Scoring method with all that colourful heatmap is utter bullshit. There is no evidence that it works and in fact there is evidence that it works in a bad way because it increases people's confidence in their risk management without actually managing risk.

More on this perhaps in Hubbard's other book called "The Failure of Risk Management", which I will have to read someday.

Insurance Premiums are good example of how rightly they quantify risk to a certain dollar value. Based on the level of risk you want to mitigate, your insurance premiums will be charged. They dont score risk in these instances as "Low" "Medium" or "High"

LETS TALK ABOUT THE REAL SHIT - MONTE CARLO

When you have a precise value of something it is easy to add, substract, multiply and divide them, but what do you do when you dont have a precise value, all you have is a range?

Lets say you want to add RANGE A variable with RANGE B variable and substract RANGE C variable

Now in Monte Carlo Simulation, you will randomly pick one value from RANGE A, one value from RANGE B and one value from RANGE C and do the operation.

A + B - C and this be repeated thousands of time. You might be wondering how is this possible? Well its possible because and when you dont know the exact values of A, B and C but you do know the probability distribution pattern of each of these individual variables.

What essentially is happening here is these distributions are being operated with each other instead of your traditional numbers, if you see the big picture. And ofcourse the resultant is also a distribution.

In fact, all risk in any project investment ultimately can be expressed by one method: the ranges of uncertainty on the costs and benefits and probabilities on events that might affect them.

Monte Carlo Simulation in Excel - Breakeven Analysis Example

Both the below videos will show to quickly utilize MCS on excel but they will only work for Normal Distributions

LCOE and Monte Carlo Simulation in Excel 2016

Introduction to Monte Carlo Simulation in Excel 2016

In a example of MCS you will find that a lot of inputs are ranges or means with std devs, and it is established what distribution these variables follow.

Giving you a range or giving you a mean with std dev and telling you the distribution is essentially the same thing.

Obviously because all previous chapters utilized 90% CI range with upper and lower bound, the author likes to use range.

However, it has to be converted to SD to be fed to Norm.inv() function in excel and for that author just divides the Range by 3.29

Apparently for Normal Distribution, Std dev = 90% CI range/3.29

Like we used Norm.inv to create Normal distribution, author has later shows how to create two more distributions on Excel, namely Bernoulli Distribution, which is sort of binary with two possible outcomes and Uniform Distribution

Bernoulli Dist created on Excel by function =if(rand()<P,1,0)

P= probability that a "1" will appear ( "0" appears with 1-P probability)

Uniform DIstribution is created on Excel by function=rand()*(UB-LB)+LB where UB LB is upper and lower bound.

Other Tools and other Resources for Monte Carlo Simulations

Some guy who is also a big proponent of MCS in business - Sam Savage and he created a tool called Insight.xla

His idea is also to appoint a Chief Probability Officer (CPO) for the firm, who manages common library of probability Distributions of various variables the firm deals with, so that more MCS can be performed.

Then Author plugs his own services for companies to be more quantitative analytical. Also featured is a long list of tools available to perform MCS

Finally the Author talks about a RISK PARADOX he has noticed in a lot of companies

ie. If an organization uses quantitative risk analysis at all in the first place, it is usually for routine operational decisions, day to day stuff.. But when it comes to larger more significant and risk management sort of decision, it hardly gets proper quantitative risk analysis.