Datafeed Toolbox includes programmatic and interactive interfaces to access financial data. Both interfaces complete the process of communicating with a financial data server in three simple steps:
The library of functions provides a consistent MATLAB programming interface for all supported financial data service providers. The functions let you customize queries to access all or selected fields from multiple securities over a specified time period. You can also retrieve intraday tick data for specified intervals and store it as time-series data.
The datafeed functions let you integrate market data into your MATLAB program or application.
Datafeed Toolbox lets you manage server connections and data requests without programming. You can access multiple data fields from a chosen equity interactively and import them into MATLAB. The equivalent MATLAB statements required for the data request are displayed in the connection history. The toolbox currently supports Bloomberg, Interactive Data, and Yahoo! Finance.
Datafeed Toolbox supports two approaches for accessing data from data service providers:
Queries are used for retrieving data that is static or a snapshot in the past, such as current or historical data. MATLAB functions such as
timeseries retrieve data from the service provider immediately after the function is used and do not update with changing market conditions.
Access Data from Thomson Reuters Datastream
Connect MATLAB® with Thomson Reuters Datastream to retrieve global financial and macroeconomic data.
Callbacks are used for retrieving data that is dynamic, event-based, or streaming in nature, such as real-time data. The procedure for retrieving real-time data is to connect to the data service, create a query, and attach a callback (a MATLAB function) that will be executed each time data is received from the service provider.
Datafeed Toolbox provides three approaches for retrieving tick data, time-series, and machine-readable news into MATLAB:
You can directly import intraday data using MATLAB functions that work with supported financial data service providers. You can directly import machine-readable news using Thomson Reuters and Bloomberg data services. The amount of intraday data and machine-readable news that can be imported varies by data service provider.
You can retrieve time-series data from kdb+ databases to the MATLAB workspace or you can write data from MATLAB to kdb+ databases. To retrieve data, you create a connection, execute kdb+ commands to perform a query, and fetch results to return them to MATLAB. To write data, you establish a connection to kdb+ and call the
insert function with the name of the table to write to from MATLAB.
Large amounts of historical intraday tick and machine-readable news can be retrieved from Thomson Reuters text file archives. Datafeed Toolbox provides two functions for importing data, in whole or in part, from the text archive files:
rdthloader– reads tick history data archive files
rnseloader– reads machine-readable news archive files
Alpha Generation Using Thomson Reuters News Sentiment and MATLAB
With the explosion of market data volumes and venues, quantitative trading firms face increasing data management complexities in their quest for alpha.
|Market Data Type|
|Data Service Provider||Real Time||Current Market||Intraday Tick or
|FactSet Data Server||✓||✓|
|Federal Reserve Economic Data (FRED)||✓|
|Kx Systems||Time-Series Database|
|Reuters Machine Readable News (NewsScope)||✓||✓||✓|
|Reuters Market Data System (RMDS)||✓||✓||✓|
|Reuters Tick History||✓|
|SIX Financial Information||✓||✓||✓|
|Thomson Reuters Datastream||✓|
|Thomson Reuters Eikon||✓||✓||✓||✓|