Automated Trading System or Algorithmic Trading System is not a new terminology for Indian Retail Traders. Popularity of ‘algos’ is caching among everyone from engineers, doctors , real estate contractors and marketing or sales executives. Many Indian Brokerage platforms started providing Algo Execution Platforms, API’s to automate traders idea.
This post gives you clarity on how to choose which decision support tools to use for automating your trading ideas. I just narrowed down the focus only to Amibroker which supports most of the features required for algo trading community.
Amibroker – Amibroker is one of the widely used trading analysis software when comes to Indian Market. It is relatively easy to build,test and deploy trading models in Amibroker than any other trading analysis software. Moreover license cost is relatively cheap compared to any other trading analysis softwares. Most of the brokers prefer supporting Amibroker , as huge community got built around Indian retail traders.
However Amibroker is a trading analysis tool and not a trading terminal. You can build your buy or sell trading logic in Amibroker. An Intermediate Bridge/API’s(mostly provided by stock broker) needs to be connected with amibroker which converts buy and sell logic into trading execution orders to the trading terminal.
Amibroker also requires realtime exchange datafeed subscription to get realtime data into charting application and many authentic datavendors like Globaldatafeeds, CDP India, Esignal supports Amibroker platform.
What are the Skills Need to Build Effective Algo Trading Solution using Amibroker
1)Knowledge in Amibroker AFL Programming, Backtesting, Optimization, System Validation of Trading Model.
2)Knowledge in understanding the Broker API’s/Bridge Functions and How to Integrate with Amibroker.
3)Knowledge about VPS servers(Virtual Private Servers)/Cloud Servers, How to remotely connect and deploy Amibroker, Bridge Components and Trading Terminal in the servers.
4)Knowledge about Setting Risk Control Parameters, Order Execution Logic).
5)Taking Data backup and Trading Logic backup at regular intervals.
6)Using statistical test like Monte Carlo analysis to determine if the trading system is broken.
Here are the some of the additional requirements to build a effective automated trading system infrastructure.
1)Good Historical data for Backtesting
2)Good Realtime data for Live Trading
3)Data Cleanliness – Data adjusted to split, bonus, dividend and Survivorship bias free data. Almost every authentic data vendor provides raw data which is not adjusted to split/bonus/corporate dividiends. In such a case one have to handle the Split/Bonus Adjustments in Amibroker manually
4)Trade Execution capabilities ( Market Order, Limit Order, Bracket Order, Cover Order)
4)Position Sizing Capabilities (Partial Profit booking, Partial Loss Booking, adding positions to existing trade (scalein) )
5)Risk Management (What to Trade, Symbol level stoploss, Portfolio level stoploss, fnoban check, Panic button squareoff..etc)
6)Good Understanding about Money Management and Types of Systems (Trend Following, Mean Reversion, Pattern Recognition, Seasonal, Cyclical)
After Taking your System Live make sure that you have the following availabilties
1)Be sure that your trading model is free from glitches and the system is tradeable.
2)There is enough time to compute and place orders.
3)Your assumptions about trading costs (brokerage commissions, slippage, taxes) are realistic.
As per exchange compliance still Retail traders are allowed to execute their trading logic in a semi automated fashion(one touch trading) and barred from fully automated trading unless getting proper approval from the exchange.
In the next section will look into the different types of execution logic and how you can use the same with Amibroker.