What could possibly go wrong?
Reminds me of Long-Term Capital Management - Wikipedia
Lot’s of very smart guys thinking their math was bullet-proof…
Take a look at his google scholar profile: Maximilian Janisch - Google Scholar
From the top 20 articles by citations, I don’t dare to read save one: [2504.08611] International Financial Markets Through 150 Years: Evaluating Stylized Facts
It looks interesting, the last sentence “robustness and generalizability” follows the path you suggest 
In the theory of financial markets, a stylized fact is a qualitative summary of a pattern in financial market data that is observed across multiple assets, asset classes and time horizons. In this article, we test a set of eleven stylized facts for financial market data. Our main contribution is to consider a broad range of geographical regions across Asia, continental Europe, and the US over a time period of 150 years, as well as two of the most traded cryptocurrencies, thus providing insights into the robustness and generalizability of commonly known stylized facts.
TBH, the guy is smart and I feel a bit like any random bitter idiot from inside paradeplatz haha
LGTM was before I was interested in this kind of stuff - but I read about it and it’s the literally first thing that popped into my head when I read that he wants to join a finance company.
My guess is he’ll end up in some crypto-blockchain-fintech incubator backed by VC and old money.
He’ll be either very rich or very rich … in experience
Yeah, he’s probably a genius. And ticks all the stereotypical genius boxes you could think of (and then some).
In any case, it won’t be the last time we read of this guy.
The guy is an asset for any startup investment fund wanting to get known and get deposits from investors. The guy is known: Mathematik-Genie Maximilian Janisch – Aus der Welt eines Hochbegabten - Reporter - Play SRF
I see the hazard of less smart and less scrupulous people using the public perception of the genius to get money from retirees while promising “higher returns”. I did not think about LTCM but of the now bankrupt Prime Energy Cleantech. They used their “ambassador” Bertrand Piccard to get deposits.
The article from him I quoted is “readable”. One of the first citations is Benoit Mandelbrot (fractals), power-law distributions. It’s the first time I see a formal presentation of “stylized facts”. I’ll finish reading later…
Sounds suspiciously like DC today…
Math is a bit like engineering. Making things works without a full understanding of how things work.
Engineers can build impressive structures without having a full understanding of materials because most of times materials works in predictable ways. But counterexamples exist: engineering failures, collapses, etc.
A stylized fact transcends changes in datasets, timeframes, measurement methods, and geographical locations; its value lies in its consistency across these dimensions, even though specific counterexamples may exist. Notable examples include Zipf’s law in linguistics, where word frequencies follow a power-law distribution…Stylized facts serve as powerful reference points for model evaluation, regardless of our understanding of the underlying mechanisms.
So, the framing of an stylized fact is to use something that is not 100% true as a building block for something else…sounds like engineering 
We need more engineers. They know the maths they use are approximations. They know the results are based on assumptions and if those assumptions are incorrect there may be trouble. They know there is uncertainty in all measurement. They normally include a large safety factor when designing.
The problem is that if the bridge falls down, the engineer takes the rap. When IT and finance systems fall over, “Biden” is to blame.
No, these days systems don’t fall over. They experience a “service degradation”.
1 Like
These stylized facts are just the outcome of a PCA, for the old fashion scientists. And PCAs are that, (Principal component analyses, for those who are not so geeky oriented) PC-- so, if the model still cannot predict at full accuracy the past data… then it is still a ‘random walk’, oder?
My experience with geniuses, is that they don’t do very well on the corporate world. But hey, best of luck to him.
1 Like
The most dangerous thing in finance is the maths. Because you are trying to model human behaviour and that never works out, but the math is good enough for a short period and it gives people a false sense of security. In the end though it always fails without exception.
The sad thinking it is that despite their intelligence, these people almost always end up believing the maths and forgetting the basic assumptions upon which their theory rests! In the case of LTCM, it was the assumptions that risk is constant and market size does not count. And for their models it was, except their time period was way too short and the market too small. Why it never dawned on them to test it with a longer time horizon and broader market, I’ll never know.
Not alone do I remember LTCM, I was involved in the clean up, as well as the MBS stuff, Washington Mutual and Lehman Brothers.
1 Like
Many a people couldn’t be less interested in seeing their work disproven. Even less so when they turn it into a product they plan to make gazillions on.