Recent content by darthtraderbeta

lol are you kidding me? Everyone believes this. The second industrial revolution is almost becoming its own category on Amazon there are so many books being wrote about it. 5 years is the difference between the Iphone 6 and Iphone X. 128 gig vs 256 gig. Peter Thiel (who also believes...

I think you are confusing hardly requires any testing with don't test at all. IMO the process should be: 1. you have a model that everything about it says it should work. 2. you test it on data a. it works and gives reasonable results inline with expectations before the test...

Well I don't know how you can disagree with something you haven't read and what you said about ML is just not correct. What you are describing is actually what people are doing who do not use ML because you can't measure what is actually contributing to the prediction like you can with ML. The...

Only thing I am really finding is this paper Warren Buffett Value Indicator vs. GDP Size ─ Is the Relationship Superlinear? https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2897317 Buffet did say this at 2900 S&P in Jun 2019 “I think stocks are ridiculously cheap if you believe ... that 3%...

I think interest rates have to distort this quite a bit though. The 10 year on 1/1/2000 was at 6% and 04/01/07 at 5%. It still feels to me that the dot com bubble has a higher valuation when you had a 6% alternative. I don't know how you would properly scale this to interest rates though. I...

At least have some balls and say black people like what you really mean you racist little bitch. You are probably still mad about getting picked last in gym class.

We are twice as far removed in time from the turtles as the turtles were from the NYSE being closed on Wednesdays to process paperwork.

If you have predictive features and the neural network is doing a supervised learning classification task then it just a classification task. To say the neural network is figuring out price action is disingenuous and a highly distorted view to someone who doesn't have any experience in ML.

To me a first principle of the market is that market regimes will change so you first have to be testing for regime change/structural breaks. To me any run of 3+ loss needs to be scrutinized that we don't have faulty assumptions in the strategy. All these statistical concepts from ergodic...

https://en.wikipedia.org/wiki/El_Farol_Bar_problem IMO it is actually a first principal of the market that most have to fail at trading in order for a small minority to gain a large profit. This is exactly what motivates trading to start with. If profit is uniformly distributed there would be...

Interesting enough, this is exactly what George Hotz on the AI podcast points out as the hardest part to getting to level 5 automation, all the counter factuals involved in the decision process. I would suspect for true level 5 automation we will need a network protocol that broadcast each...

How the fuck can someone digest Yuval's work and come away with the idea he his a doomsayer? That is a ridiculous characterization.

I wouldn't put a huge bet on it but I could see close to the day of reckoning USDT market cap pushing ETH market cap for #2.

I actually have very little interest in RL and neural networks although my avatar was made by a GAN. mondrian random forest, causal forests, ensemble methods in general for regression are what interest me.

There is a huge difference between learning to be a software engineer vs learning data science. Data science is pretty easy. I personally hate code and want to write as little as I can. I am interested in results and analysis, not code. Learning to use pandas in a jupyter notebook is not...