Predatory High Frequency Trading Machine Learning Thread

[Esterincertis]

quote College for programming(C ). My highschool sweetheart went to Bryant University at the US to get Actuarial Mathematics and graduated only shy of Suma Cum Laude. I was still helping her reason her way through a few assignments until she graduated(I had a passion for math). I worked through all this justifiion for over a decade to the explicit intention of doing everything you've spoken of BEFORE I put a single trade. When I began studying plogy, the issue came to light. The answer came from quantum mechanics. At the end of the day...

Also I addressed the hardware issue, using cloud based vms I might acces lot

[Rororo93]

quote I can't work at two sigma since Iam seventeen and have not finished highschool. As for black swans, the predatory egies find 30-40 opportunities a month each pattern recognition model. You dont have to be right all the time with neural networks, that I quantify the sign against price for a correlation coefficient. Thus the versions always presume some error. The pc jus keeps detecting better and better versions. It likely will not be ideal, and doesnt have to be. Just good enough to net a profit.

That's dull. The cash is unbelievably simple to acquire. Figuring out how to get it's always the interesting part. Figuring out exactly what you need to do with it maybe not so much. It leads to quite a bit of discontent. I wish you the very best of luck.

Also, I highly doubt anyone who said 'No' is pleased with their own lives. understands.

[Tataylo]

Hi, the issue is that you are searching for patterns in place of fractal data ( https://en.wikipedia.org/wiki/fractal).
But in fractal data you will find by definition no patterns.

Here a nice TED talk about HFT

[staticbangs]

A fractal is a pattern!!!!

[Rororo93]

Hi, the problem is that you're interested in finding patterns in set of fractal data (https://en.wikipedia.org/wiki/fractal). However, in fractal information you will find by definition no patterns. Here a nice TED discuss HFT https://youtu.be/fawD7Nj74NA

Excellent video. Perhaps some new traders will find it enlightening.

[Rororo93]

quote Also I already addressed that the hardware problem, using cloud established vms I can acces lot

Try approaching it from another angle. The hardware problem is not the issue. The input is the issue. The information is tainted.

[Esterincertis]

I'm starting implementation. How are scalpers/hftraders doing today?

[Esterincertis]

Tested it, and it created it's first commerce, netted 60 pips on the peso, but had 2 losing trades and just one winner.

[Dextroxs]

quote Try approaching it from a different angle. The hardware problem is not the issue. The input is the issue. The data is already corrupted.

The data is definitely corrupted return to Greece having a ghost deal with Goldman sachs for a currency exchange which made them look 3 billion dollars better off from becoming 2 for 1 euros on the market, while they secretly owe Gold guy and then no one gets in trouble. Logic would say that gold guy has black deals with every other nation and so no one will rock the boat as they're in precisely the same boat. Meaning the exchange rate is seriously out, because these deals just half is about the exchange another half is unknown. So than the market is presently carrying skewed sovereign risk, imagine if all Euro nations owe large sums of cash to the bank off the books, yet have healthy current accounts because they bank poured liquidity into their market/currency for a mortgage. (leading to the conclusion we want more data, lots more because discovering black deal cash would be discovering the error values at a proper fundamental regression on the market, random walk with a drift, least squared linear regression or so on)

[Dextroxs]

JPY I believe is the main supply of black bargain swaps being exceptionally cheap and quite liquid.