Introducing Rblpapi

Dirk Eddelbuettel

2015-08-13

Introduction

The Rblpapi package connects the R language and environment for statistical computing to the Bloomberg services supported by the official Bloomberg APIs.

The Rblpapi package is provided in both source and binary (for Windows and OS X) via the CRAN network for R. Source code is available at the corresponding GitHub repo; the blp repo contains the Bloomberg API components required during the build.

Usage

The next few section illustrate key functions within the package. All functions also have proper help pages for fuller details.

Package Load

Use

to load the package. You can also set options which will automatically connect at package load; see the next section.

Connecting

Use

which will connect to the Bloomberg backend. Default values for the IP address (127.0.0.1) and port (8194) are used and can be overridden both as function arguments and via global options blpHost and blpPort. Moreover, if option blpAutoConnect is set to TRUE, this connection is done at package load.

The resulting connection object is stored in a per-package environment providing a default value. If needed it can be overridden in each accessor function.

As the connection object is being held onto, the blpDisconnect() function is implemented as an empty function without side-effects. The internal function defaultConnection() returns the current connection object.

bdp: Bloomberg Data Point

Current (or most recent) values of market-related instruments can be accessed via the bdp() function:

bds: Bloomberg Data Set

The bds() function can retrieve data of a more static nature:

bdh: Bloomberg Data History

Historical data (at a daily granularity) can be retrieved with bdh():

getBars: OHLCV Aggregates

The getBars() function retrieves aggregated Open / High / Low / Close / Volume data. For example, in

the default values of six hourly bars for the lead ES future.

getTicks: Transactional Tick Data

The getTicks() function retrieves tickdata, albeit with timestamps at a minute granularity. For example, the call

retrieves all ticks for the given security over the last hour.

fieldSearch: Query For Fields

The fieldSearch() helper function can search for available fields which can be used with the bdp(), bdh() or bds() functions. For example,

search for fields describing volume-weighted average price fields

beqs: Bloomberg EQS Queries

The beqs() function (which was contribued by Rademeyer Vermaak) can access EQS functionality:

Acknowledgements

Most valuable contributions from both Jonathan Owen (who kick-started builds on Windows when we considered these to be impossible) and Jeroen Ooms (who took care of builds on OS X) are gratefully acknowledged.