Python

Access Pacific data with the most popular data science solution using the pandasdmx library

Overview

PDH.stat is part of the sdmx Python package developed by Paul Natsuo Kishimoto. More information about the package can be found here.

Install sdmx

These steps have been tested with Python 3.7.4 in an Anaconda environment on Windows 10.

Start by installing sdmx. To learn more about the package, see the code here.

To install with pip from the command prompt (note: include the '1' at the end of sdmx): pip install sdmx1

In a Python project, import the package: import sdmx

Connect to PDH.stat

To see the available sources and find PDH.stat:

import sdmx
sdmx.list_sources()

The source abbreviation for PDH.stat is "SPC" (Pacific Community), as shown below.

To connect to PDH.stat and then view its available data flows:

spc = sdmx.Client('SPC')
datasets = sdmx.to_pandas(spc.dataflow().dataflow)
datasets

To connect to a data flow and convert it into a pandas Multi-index series:

data = spc.data('DF_CPI')
df = sdmx.to_pandas(data)
df

And then to turn the series into a dataframe, reset the index:

df = df.reset_index()
df

For an example of how to use the plugin in combination with the API key and parameter settings, see the time series plot example.