Broker Operations

Relevant source files

The following files were used as context for generating this wiki page:

• hyperdrive/Broker.py
• hyperdrive/Algotrader.py
• hyperdrive/Exchange.py
• scripts/update_symbols.py
• scripts/execute_order.py
• test/test_Broker.py

Introduction

Broker Operations in the hyperdrive library encompass the interfaces and logic required to interact with financial intermediaries and exchanges. This system provides a unified way to authenticate with services, retrieve historical market data, manage portfolio symbols, and execute trades. The primary implementation currently focuses on Robinhood for portfolio management and Binance for cryptocurrency exchange operations.

The module facilitates the transition from quantitative research to live execution by providing standardized classes like Robinhood and Binance that abstract the underlying API complexities. These operations are critical for maintaining up-to-date symbol lists and executing orders based on signals generated by the library's research components.

Sources: hyperdrive/Broker.py, hyperdrive/Exchange.py, hyperdrive/Algotrader.py

Robinhood Implementation

The Robinhood class provides the core functionality for traditional equity broker operations. It handles Multi-Factor Authentication (MFA) and provides methods to query holdings and historical data.

Authentication and Initialization

Initialization requires a username, password, and MFA code. If not provided directly, the class retrieves these from environment variables (RH_USERNAME, RH_PASSWORD, RH_2FA).

Key Functions

Function	Description
get_hists	Retrieves historical data for a list of symbols with specified span and interval.
get_names	Resolves stock symbols to their respective company names.
save_symbols	Persists all symbols in the current portfolio to a CSV file.
get_holdings	Fetches and builds a dictionary of current account holdings.

Sources: hyperdrive/Broker.py:38-84

Data Flow for Symbol Management

The library uses automated scripts to sync local data with the broker state. The save_symbols method retrieves active holdings and saves them using the PathFinder utility.

Exchange Operations (Binance)

While Broker.py handles portfolio management, Exchange.py focuses on execution logic specifically for the Binance exchange.

Order Execution Logic

The Binance.order method calculates quantities based on a spend_ratio and account balance. It automatically adjusts for trading fees (0.1%) and rounds quantities to meet the exchange's stepSize or asset precision requirements.

Configuration Options

Parameter	Type	Default	Description
testnet	bool	False	Whether to use the Binance Testnet environment.
spend_ratio	float	1	Percentage of available balance to use (0.0 to 1.0).
side	str	N/A	Must be 'BUY' or 'SELL'.

Sources: hyperdrive/Exchange.py:10-25, scripts/execute_order.py:24-28

Integration with Algotrader

The HyperDrive class in Algotrader.py serves as a high-level wrapper that composes these broker operations into a single interface. It initializes a Robinhood instance and provides an option to load the current portfolio upon instantiation.

Sources: hyperdrive/Algotrader.py:4-16

Conclusion

Broker operations in hyperdrive bridge the gap between data analysis and market interaction. By utilizing specialized classes for different providers (Robinhood and Binance), the system maintains a clean separation between portfolio tracking and trade execution. The inclusion of automated scripts for symbol updates and order execution ensures that the local environment remains synchronized with the live brokerage state.

Sources: hyperdrive/Broker.py, hyperdrive/Exchange.py, scripts/execute_order.py