fbpx

Automated trading tips from Trading Show NYC

(Last Updated On: October 21, 2015)

Automated trading tips from Trading Show NYC

Big thanks to the NYC Contact for sending this

The Trading Show NYC 2015 Summary:

This was the best trading show out of the three I went to!

——————————————————-
The Triple Crown of Technology

Once basic infrastructure is in place
the focus is on getting as close to zero as possible

1) Zero Cost
2) Zero Latency
3) Zero Time to Market
—————————————————–
Trade Metrics

1) P/L + Max Drawdown
2) Cash Flows from long term trading activity
3) Discounted Cash Flows from long term trading activity

—————————————————–
Tiered Data Pipeline

L1 Cache -> L2 Cache -> L3 Cache -> Memory -> SSDs -> Spinning Disks

Avoid backup in Data Pipline by using 8K circular queues that are cleaned at the speed necessary
to keep them empty (using multi-threaded cleanup code)

XPoint Tech as the next high speed storage after DRAM Flash (PCI-V3.0 SSD)

Merchant Silicon (Run on bare metal switches) vs x86 Architecture (Run on OS)

Speed Improvement in separation of concerns techniques

VMs -> Containers -> Binaries

AMPS -> Streaming DB with Filtering Capabilities

Kafka -> Subscriber Model/No Filtering

—————————————
Kernel Bypass Technique:

Turn UserSpace Kernel into Packet Processor
to accelerate overall networking speed.

Solarfare:
http://www.solarflare.com/
—————————————

Tick Timestamp Issues:

1) Time Granularity
2) Time Synchronization across different markets

—————————————–

Trading Strategy Evaluation:

1) Does the algorithm mirror the trading strategy?

2) Does it minimize trading cost?

3) Max Alpha

4) P/L Stability

—————————————–

Quantum Computing:
1) 2^N Total Solution Space where N is the total number of qbits

2) Superposition of the QBit (bit is both zero and one at the same time)

3) Quantum Entanglement -> Probability Distribution describes
how many and which qbits are affected.

4) Quantum Tunneling -> Can tunnel through 3D Landscape or Volatility Surfaces
1,000 qbit CPU yields 2^1000 solution space
Multiple parameter optimization problem must first be converted to a
binary optimization problem before the Quantum Computer can solve it.

Brute force simple binary algorithm which is great for differential evolution optimization
algorithms.

http://www.solarflare.com/

 

Join my FREE newsletter to see if we implement any of this

NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!

Subscribe For Latest Updates

Sign up to best of business news, informed analysis and opinions on what matters to you.
Invalid email address
We promise not to spam you. You can unsubscribe at any time.

NOTE!

Check NEW site on stock forex and ETF analysis and automation

Scroll to Top