About This Event
“Turn raw data into official statistics faster, without cutting corners.”
A five-day virtual regional workshop introducing foundational data-science skills applied to macroeconomic statistics. You will use Python and R to clean, reshape and validate large datasets, automate routine compilation, explore new data sources such as web-scraped prices and administrative records, and visualise results clearly.
What You'll Explore
Reusable Python and R scripts for cleaning and validating macro data
An automated, reproducible pipeline for one of your own datasets
A practical approach to assessing web-scraped and administrative data
Chart and dashboard templates for presenting statistical releases
A plan for scaling data-science methods across your department
Who Should Attend
Open to all qualifying staff, particularly: Statistics Compilers, Data Analysts, Research Economists, IT & Data Management Staff, Balance of Payments and Monetary Statisticians.
Why This Course Matters
Automated Compilation
Scripts replace copy-and-paste spreadsheet routines that are slow and error-prone. Automation frees compilers to spend their time checking and explaining numbers rather than assembling them.
New Data Sources
Web-scraped prices, payments data and administrative records can fill gaps in traditional surveys. Knowing how to handle them lets your department produce timelier indicators at lower cost.
Reproducible Quality
Code-based workflows leave a clear audit trail from source data to published table. That reproducibility makes quality checks easier and your statistics more defensible when users ask questions.
Programme
Day 1
Data-science foundations for official statisticians
You will set up a working environment in Python and R, learn the core data structures and notebooks, and see how data-science workflows fit into the statistical business process.
Day 2
Cleaning, reshaping & validating macro datasets
You will import, clean and reshape real monetary, price and trade datasets with pandas and tidyverse tools, and write validation rules that catch errors before they reach publication.
Day 3
Automating compilation & reproducible pipelines
You will turn a manual spreadsheet routine into a scripted, reproducible pipeline with version control, so the same inputs always produce the same outputs and every change is traceable.
Day 4
New data sources: web scraping & administrative data
You will collect web-scraped price data, work with administrative and payments records, and assess their quality, coverage and representativeness against the needs of official statistics.
Day 5
Visualisation, SDMX exchange & a capstone project
You will build clear charts and dashboards, exchange data in SDMX format, and complete a capstone project applying the week's tools to a dataset from your own department.
Standards & Faculty Benchmark
UN Fundamental Principles of Official Statistics
The UN principles on professional independence, quality and confidentiality in official statistics.
IMF Data Quality Assessment Framework (DQAF)
The IMF framework for assessing the accuracy, reliability and soundness of macroeconomic statistics.
SDMX standards
The Statistical Data and Metadata eXchange standards used by the IMF, BIS and others for data transmission.
New tools only help if the statistics they produce meet the same quality and confidentiality standards as before. Anchoring your data-science work to the UN principles, DQAF and SDMX keeps innovation within the rules your users and the IMF already expect.
Is This Right for You?
- ☑You compile macroeconomic statistics using spreadsheets today
- ☑Your institution wants to modernise its statistical production
- ☑You want practical coding skills without a computer science degree
Good to Know
No programming experience is required; the workshop starts from first principles and uses ready-made notebooks. Delivered live online with guided labs. You leave with reusable Python and R scripts, an automated pipeline for one of your own datasets and a plan for scaling up.
The Bottom Line
Return with working code, an automated pipeline and the confidence to bring data-science methods into your statistics work.
Recommended For
Open to all qualifying staff, particularly: Statistics Compilers, Data Analysts, Research Economists, IT & Data Management Staff, Balance of Payments and Monetary Statisticians.