Make better decisions through Probability & Statistics _ intro
Updated: May 8, 2020
I am an engineer, I transitioned from mechanical engineering to financial engineering; while the former is the one I love the most, the one I am always tinkering with and the one I am going to go back eventually, the latter is the one I consider to be a turning point in my way of thinking.
The reason why I think my financial studies were a decisive point is not that they gave me the tools to get most of my decisions right, quite contrarily, it is because they made me doubt about almost all my conclusions. I think also that trading and focusing on financial derivatives helped a lot -- financial derivatives are extremely real things; you often either quit after heavy losses or, after the first red flags, you immediately admit your shortcomings, regroup and start over through little steps.
What was enlightening to me within the financial engineering course was not corporate finance, nor investment banking, nor private equity; it was risk management – let’s reduce here risk management to probability and statistics. Please note, even though I was introduced to the argument through finance, I think the real value is the possibility to transfer that approach to any personal and professional decision.
I was very fortunate in approaching probability and statistics through financial engineering, that’s mainly because probability distributions within financial markets may be among the toughest ones to handle in our common professional domains. In other environments, distributions may be more prone to being described and controlled through our math and models – that’s likely why we continue to statistically improve in manufacturing and aeronautics while we keep having the same blackouts in finance.
Financial markets being not an easy application was exactly the reason why I think it was the best possible domain to start from. Very often, tough problems push you to stress concepts constantly, even though actual applications may then come after long time and in different domains.
There is more about my feelings on probability and statistics, I do not feel like I can handle much with respect to the subject. Honestly, I think I’ve got barely right what I tested, simulated and applied. Just an example of my shortcomings: I like to think in terms of Pareto distribution, very frequent in financial applications – which many may know for the 80-20 rule – however, anytime I start feeling like I know all its important features, I discover a new peculiarity – maybe through a wrong trade, a simulation in Python or a comparison with a real-life scenario.
This post is meant to be an introduction to the following series, that I hope you’ll find useful as much as I did; it will be made of three parts that are the three main moments in my relationship with probability and statistics so far. These three arguments and my main takeaways will be the followings:
1. Bayesian probability made me realize that in any scenario we are more likely to be wrong in our initial hypotheses than right. We can make better decisions only by trying to disproof ourselves rather than validate our initial thinking.
2. Everything starts from the assumed distribution of the data; if we start and stay with the wrong one, we’ll likely never get it right.
3. The best decisions we can make are those based on ranges rather than numbers; our success may depend on that.
While hoping you’ll follow me in this tale of my three main turning points, I also hope I won’t find in the meantime other disproofs to my understanding of them, otherwise I’ll have to go over the drafts of its parts again. Seriously, my objective is just to make the reader curious about topics that were extremely interesting and useful to me. As already said, I am personally in my quest to get right as much as I can about the subject, therefore I am in no position to teach anybody – I think this applies to any domain I focus on, not only probability and statistics.
Probably, even experts and professors should emphasize the fact that they are often giving us just experiences and point of views rather than definitive answers – this is what happened with professors I consider the best I've had. Ultimately, maybe, only personal research and practice can really teach us something.
See you in part-1
Main image's tag: