nyc algo meetup_jan_16_14

Download Nyc algo meetup_jan_16_14

If you can't read please download the document

Upload: twiecki

Post on 16-Apr-2017

1.004 views

Category:

Economy & Finance


0 download

TRANSCRIPT

Algorithmic Trading in Python

Thomas Wiecki@twiecki

About me

About me

Rise of the machines
84% of all stock trades are done by computers

vs

RationalitySpeedCapacityIntuitionUnderstandingSpeech

- Cognitively biased / emotional- Human- Limited+ 50 years of experience+ Knows how the market works+ Native speaker

+ Purely analytical+ 1 TFLOP+ Big Data- None- Depends on programming- I'll be back

Pythonic Algorithmic Trading Pipeline

Data AnalysisGenerate ideas

Tools:Pandas

Matplotlib (Seaborn)

IPython

BacktestingTest on historicaldata whether ideacan be exploited.

Tools:ZiplineQuantopian

PyAlgoTrader

Forward testingPaper-trade onreal-time marketdata.

Tools:Quantopian

Live tradingTrade with realmoney.

Tools:Quantopian (soon)

optimize

Algorithmic Trading: Backtesting

Historical stockdatadatetime

Price

Volume

Trading AlgorithmApplies transforms (e.g. mavg)

Makes trading decision

Places ordersTransaction costs

Slippage

Order delay

Performance and Risk analysis

Zipline

Trading simulator/backtester written in Python

Open-Source (Apache 2.0)

Stream-based (prevents look-ahead bias)

Batteries includedMoving average, Sharpe, alpha, beta...

Used in production on Quantopian.comContribute back to community

Linus' law: "given enough eyeballs, all bugs are shallow"

http://zipline.io

Interoperability

Architecture

Go to: http://nbviewer.ipython.org/4631031/

Conclusions

Zipline gives you full control over you backtest.

Copy & paste your Zipline algorithm to Quantopian:Share it, get feedback.

Test it on minute data.

Forward test it.

Go live (soon!)

Come hack Wall Street.

Questions?

@twiecki@quantopian