That reminds me of Scott Aaronson's post about the early signs a complexity paper is unlikely to be valid: https://www.scottaaronson.com/blog/?p=304 Not tex is the first point.
Ergodicity becomes a problem here. The nonlinear systems typically studied by complexity and chaos theorists have strong fixed rules that do not change in time. Markets have systematic and structural changes which can make prior observations completely irrelevant.
Surely that title is backwards. I really doubt there is any ML methods that are inspired or based on econometric approaches, but there is no doubt that there exist econometric problems which can be approached using machine learning.
Just so you know, economic regression modelling (the main application econometrics) has existed since the late 1800's. ML has existed since the 1950's. So yes, economists, econometricians and mathematicians who have worked on applied economics would have inspired many aspects of ML, and vice versa.
So he's willing to concede space on the scientific landscape in the face of increasingly oppressive 'political correctness'?
Don't research something if it may hurt someones feelings! Great.
A few years ago I graduated with a PhD in statistics with lots of ML inspiration. Since then I have always dreamed of applying my knowledge and skill in this domain. However, despite the belief I was 'probably' in a decent position to do so, I consistently read about how impossible it was. I have a boring 'normal' persons job, but, posts like this are somewhat reassuring that I made a reasonable decision to abandon a life of fruitless datamining and overfitting.
I am keen to second here. With a PhD in probability & loads of experience in data analytics, my experience has told me that we are too ignorant and sometime too ambitious to try to predict a the outcome of a stochastic process (e.g. financial time series) without knowing that the amount of information required to make a sound prediction is far beyond those we have. Unless there's a very clear dominating signal among thousands of information sources, very often we are trading on noise.
Although I don't necessarily agree with all the points in this article, it just reminds me what Poincaré said:
`You ask me to predict for you the phenomena about to happen. If, unluckily, I knew the laws of these phenomena I could make the prediction only by inextricable calculations and would have to renounce attempting to answer you; but as I have the good fortune not to know them, I will answer you at once. And what is most surprising, my answer will be right.'
-- Poincaré, H. (1913) The Foundations of Science. New York, The Science Press. p. 396.
I don't think the message here is "don't do it," but "have domain knowledge." The crux of the paper was scientists applying ML to a bunch of data without really understanding trading.
You can actually have scientists find signals in data they have no domain experience in. In a typical hedge fund the quantitative researchers will be a different group from the quantitative developers and traders. There are fuzzy lines between those depending on culture, but those three groups are broadly the front office. You really need domain experience for execution and risk management, but pure insights can be derived without necessarily needing any domain experience.
That said, quant researchers typically understand how the market works. They are just able to quickly excel without a background in it.
It's easy to forget that this is a highly competitive field.
You're used to see the techniques you work with capture signal because there isn't an army of PhDs in math, physics, and computer science working around the clock to trade any signal out of that data.
In the end, it doesn't even matter if you're the best statistician in the world: whatever signal you detect may simply not be worth the effort you put into detecting it.
I have colleagues that end up writing the same god damn code over and over again because they fail to appreciate the idea of 'generic' code and 'frameworks'.
If you're interested in Security Now, you might like Risky Business: https://risky.biz
It feels like an upgrade to me, by people who spend their days working in infosec. They've had interviews with members of LulzSec, the NSA's General Counsel, and the guy behind PwnAllTheThings. They also broke the inside story of what actually happened in Australia's national census outage.
My one hesitation is it's starting to feel like pro-5eyes propaganda. They were very dismissive of the Shadow Brokers last year (because "NSA superiority"), and have had to rewrite history since the actual 0-days & WannaCry were released. So I don't listen to every episode anymore, but I do find it informative.
Rationally Speaking and Security Now in the same list make my head hurt. Julia Galef has entertaining, highly intelligent and nuanced conversations with brilliant people. Gibson tries to sell tools that are "good because they are written in assembly".
Yes and no. I mean the demo is great, but can I actually have one? No? Well then I can only be so impressed.
I know, they're coming soon, but I've been waiting for years and at some point the enthusiasm wanes when it's been right around the corner for several years.
Hey, maybe they're trying to solve a problem that's harder than anyone expected.
