CSO Ireland Data

==============================

Easily download data from the CSO PxStat API as Pandas datasets.

Uses requests-cache for super fast access to cached requests and easy persistence with multiple storage backends.

Installation

To install, just use pip:

pip install cso-ireland-data

Usage

Getting started

First, set up a CSODataSession.

• By default, this is really simple.

  from cso_ireland_data import CSODataSession
  cso = CSODataSession()
  

• If you want to add caching, no problem! All the functionality of the requests-cache package is available through cached_session_params.

  from datetime import timedelta
  from cso_ireland_data import CSODataSession
  
  cso = CSODataSession(
      cached_session_params={
          "use_cache_dir": True,  # Save files in the default user cache dir
          "cache_control": True,  # Use Cache-Control response headers for expiration, if available
          "expire_after": timedelta(days=1),  # Otherwise expire responses after one day
      }
  )
  

• Stuck behind a corporate firewall that causes SSL certificate issues? Also no problem! All the functionality of the requests get() method is available through request_params.

  from cso_ireland+data import CSODataSession
  
  # Tell requests.get() it's ok not to verify SSL certificates when getting data.
  # !!! Only do this if you're absolutely sure it's what you need !!!
  cso = CSODataSession(request_params={"verify": False})
  

Getting the data catalogue

To get a catalogue (Table of Contents) of all the datasets that are available through the API, use get_toc().

NB Requests for the ToC sometimes time out on the CSO API. IF this happens, try again!

cso.get_toc()

Getting a table using its ID code

To get the whole contents of a particular table hosted on the Statbank API, use get_table().

You just need to know the ID code of the table, which you can look up using get_toc().

wpm29 = cso.get_table("WPM29")
wpm29.head()

Getting some common tables quickly

The CSODataSession class includes some useful methods to get data from commonly accessed tables quickly.

Monthly Consumer Price Index (CPI)

By default, the monthly_cpi() method returns a single column corresponding to the 'All items' headline CPI in the source table.

Also by default, this index is re-normalized to the most recent month - you can toggle this by setting normalize_to_most_recent to False.

simple_cpi = cso.monthly_cpi()
simple_cpi.tail()

It's also possible to pass a list of commodity groups:

commmodity_group_cpi = cso.monthly_cpi(
    commodity_groups=[
        "All items",
        "Alcoholic beverages and tobacco",
        "Health",
        "Recreation and culture",
    ]
)
commodity_group_cpi.tail()

Live Register

Use the live_register() method to get Live Register numbers (optionally broken down by Age Group and Sex) by month. This is a long data series, starting in April 1967 and still continuing every month, so it may be convenient to specify a start and/or end date for the data returned.

The Live Register data series is based on a monthly point-in-time count of people who have active Jobseeker claims with the Department of Social Protection (DSP), and these counts are extracted from DSP's administrative computer systems on a particular day every month.

Because of this, live_register() returns three possibly useful dates for each month:

1. 'Month' is the index of the data frame. It's just the last day of each calendar month.
2. 'reference_date' is the date of the point-in-time count of people with active Jobseeker claims. It's the last Friday of each month before May 2015, and the last Thursday of the month from then on.
3. 'extract_date' is the date on which the source administrative data was actually extracted - it's always the Sunday after the reporting_date.

live_register = cso.live_register(start=datetime(2010, 1, 1))

Life Tables

The life_table() method by default returns a complete life table for the most recent source data vintage.

life_table = cso.life_table()
life_table.head()