Predatory High Frequency Trading Machine Learning Thread | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Algorithmic Quant Trading (Machine Learning + Stat-Arb) 25 replies

Machine Learning Trading Algos 10 replies

Trading and machine learning 4 replies

My most recent advancements into machine learning 16 replies

Predatory and Parasitic Trading 36 replies

Attachments: Predatory High Frequency Trading Machine Learning Thread
Exit Attachments

Predatory High Frequency Trading Machine Learning Thread

  • #21
  • May 16, 2017 9:30pm May 16, 2017 9:30pm
JPY I believe is the main source of black deal swaps being extremely cheap and very liquid.
  • #22
  • May 18, 2017 2:13pm May 18, 2017 2:13pm
tested my hft strategy for three days on limited capital+volatility allocation just to find out it I forgot to multiply my dynamic allocacation distribution by -1. lost .3% of return on trade size, but just restarted it, it's running and it's netting profit

AWESOME
Quant Trader - My Blog: quantstop.blogspot (dot) com
  • #23
  • May 18, 2017 2:24pm May 18, 2017 2:24pm
Quoting exofabulous
black deal swaps
I am familiar with fx/cross currency swap contracts, but not black deal. What distinguishes them?
Quant Trader - My Blog: quantstop.blogspot (dot) com
  • #24
  • May 21, 2017 12:43pm May 21, 2017 12:43pm
Quoting aroxalot
tested my hft strategy for three days on limited capital+volatility allocation just to find out it I forgot to multiply my dynamic allocacation distribution by -1. lost .3% of return on trade size, but just restarted it, it's running and it's netting profit AWESOME
so there is no fundamentals in this system??
  • #25
  • May 21, 2017 4:05pm May 21, 2017 4:05pm
Quoting Neo1
{quote} so there is no fundamentals in this system??
there is fundamentals in the sense it analyzes transactions/volume rather than price. Later I plan to add news events and proximity to them and since them as a input as well.
Quant Trader - My Blog: quantstop.blogspot (dot) com
  • #26
  • May 21, 2017 6:47pm May 21, 2017 6:47pm
Quoting aroxalot
{quote} there is fundamentals in the sense it analyzes transactions/volume rather than price. Later I plan to add news events and proximity to them and since them as a input as well.
is it running ? what's new about it can you give some details
  • #27
  • May 21, 2017 7:32pm May 21, 2017 7:32pm
Quoting Neo1
{quote} is it running ? what's new about it can you give some details
it's inactive, no large enough transactions to make a profit off its effect on the market. I've noticed it trades alot around periods of high volatility.
Quant Trader - My Blog: quantstop.blogspot (dot) com
  • #28
  • May 21, 2017 7:36pm May 21, 2017 7:36pm
Quoting aroxalot
{quote} it's inactive, no large enough transactions to make a profit off its effect on the market. I've noticed it trades alot around periods of high volatility.
dont you think that the relations the software found are only available because of the small amount of data??
  • #29
  • May 21, 2017 7:53pm May 21, 2017 7:53pm
Quoting Neo1
{quote} dont you think that the relations the software found are only available because of the small amount of data??
i used a in sample test, but later tried out of sample. found great success on the out of sample data. not likely the result of data mining bias.
Quant Trader - My Blog: quantstop.blogspot (dot) com
  • #30
  • May 21, 2017 8:27pm May 21, 2017 8:27pm
Quoting aroxalot
{quote} i used a in sample test, but later tried out of sample. found great success on the out of sample data. not likely the result of data mining bias.
how much data for both??
  • #31
  • Jul 29, 2017 10:35am Jul 29, 2017 10:35am
Quoting aroxalot
{quote} i used a in sample test, but later tried out of sample. found great success on the out of sample data. not likely the result of data mining bias.
Hi I'm over 3 times your age and please forgive me for saying that I'm totally lost as far as understanding what you're doing except that the part about predicting big player/large investor moves sounds incredibly interesting.

Anyway what I did notice is that you are one of a very few who saw and traded the big drop yesterday. Was that the software or luck? If it was the software you have a disciple!
  • #32
  • Last Post: Jul 29, 2017 5:35pm Jul 29, 2017 5:35pm
Is anyone looking for machine learning projects I have one that I could use some help on!
0 traders viewing now