If you generate a process using a Cauchy distribution, you can observe finite error bars, but actually the variance of the generating process is infinite.

No matter how hard you model you won’t be able to predict processes that are fundamentally unpredictable. And you would get fooled because you only observe finite amount of data.

It is surprising how complicated the math gets even if you try to model very simple processes (eg think of the n-body problem and how complexity increases with every addition of a body). It is not a given that complicated models mean you’re modelling a complicated process.

You are correct.

But when people say "finance is science" what they usually mean is "here is the complicated math that proves you can't lose money on this, we've modeled everything".

As the joke goes, 6 sigma events happen in finance every week.

> when people say "finance is science" what they usually mean is "here is the complicated math that proves you can't lose money on this, we've modeled everything"

100% agree. When I was an algorithmic derivatives trader, we joked that the math was there to scare up investors and scare off compliance. Little did we know...

> we've modeled everything

That's a great science joke.

This is silly.

This is like saying physics is not science because the nutjobs claiming we will recieve divine revelation by praying to "the quantum energy field" use physics-y terms in their BS.

Re: "The error bars are not infinite."

I feel like they are infinite? Because, for example, in hyperinflation, there's no upper bound on how long someone is going to keep printing money.

Any given instance will stop at a finite number, but you can't bank of that being the high water mark.

Error bars aren't limits, they're percentiles.

Of course that can happen. But it's a low-risk event that falls outside the, say, 99% confidence interval.

It's like the notion of a 100-year flood. Of course there could be a tsunami or a dam failure that completely inundates an entire city, but at some point you've got to accept a small risk and ensure you are covered for it.

> If you follow the scientific method, it's science.

OP's claim was that quantitative finance was *hard science*. Requirements regarding predictability are way more stringent than merely observing stuff and seeing how it responds to an input.

> OP's claim was that quantitative finance was hard science*

These are colloquial terms [1]. We might as well argue about whether Pop Tarts are ravioli or tacos sandwiches.

[1] https://en.wikipedia.org/wiki/Hard_and_soft_science

What is(n't) a taco is a semi-solved problem[0].

[0]: https://cuberule.com/

> These are colloquial terms [1].

Colloquial terms whose concrete meaning does not correspond to OP's claim.

There is no ambiguity in this: if your models are not testable and fail to predict behavior then it's not hard science.