Skip to content

Latest commit

 

History

History
executable file
·
50 lines (34 loc) · 2.08 KB

README.md

File metadata and controls

executable file
·
50 lines (34 loc) · 2.08 KB
Raw

quantdb

Maintains and updates a PostgreSQL database of stock market data gathered with the quantmod package for R.

Schema structure follows SQL industry standards with separate layers for loading raw CSVs to an external table (ss_ext.ext_price), casting into a staging table (ss_stg.stg_price), and loading into a target table (ss.price). More detailed company data are also loaded into another table (ss.info) ready to be joined with the ss.price table, e.g.:

SELECT * FROM ss.price
  AS t1
INNER JOIN ss.info
  AS t2 
  ON t1.symbol = t2.symbol
;

Install dependencies

Requires an installation of R and PostgreSQL, and R packages dplyr and quantmod R package:

require(dplyr)
require(devtools)
devtools::install_github("joshuaulrich/quantmod")

Set parameters in ./config.sh

Set global parameters in './config.sh', including directory database parameters (host, user, database, password), start date to return stock data from, and a list of ticker symbols you're interested in getting data for. See ./R/build_my_symbols_ex.R script for an example of getting symbols using quantmod and my_symbols.txt for an example file.

Initial database build using ./buildDB.sh

After setting global parameters, build a new database (or overwrite an existing database) using:

./buildDB.sh

Update database using ./updateDB.sh

Once a database has been created, you can update it using:

./updateDB.sh

This could be automated e.g. weekly with a job scheduler.

Contact
  • website / blog: jogall.github.io
  • email: joedgallagher [at] gmail [dot] com
  • twitter: @joedgallagher