Harmonize is a time series management and visualization system designed to make complex time series data accessible to users without programming skills, while also helping management and analysts save time.

The system stores time series data in a generic and flexible format, allowing easy comparison of datasets with different frequencies, base years, and aggregation levels. Harmonize is based on Microsoft SQL Server and can be accessed through a Windows C# application or programmatically via Python and R APIs.

Harmonize is designed for organizations that need to store, share, and analyze time series data across multiple domains. Typical use cases include data related to environment, prices, health, employment, power, energy, and other statistical or operational datasets.

The system is especially well suited for organizations that want to centralize data from different sources, make it accessible to users with varying technical skills, and enable consistent analysis and comparison across domains and time frequencies.

• Microsoft SQL Server
• Database model: Harmonize.sql
• Backup file: Harmonize.bak
  These ensure the Harmonize database is available and correctly configured.

• Windows C# application for visualization and interaction
• Requires Visual Studio to compile and build
• Not required, but significantly improves usability

• Python installation required for API interaction
• Python samples are available in the Python/ directory
• Used for:
  • Uploading / upserting new datasets
  • Extracting data for further calculations (e.g., seasonal adjustment)
  • Writing processed data back into the database
• Similar workflows apply for R

-- Stored procedures available from Python / R and MSSQL Database

• Create loadset
• PutTime :Update or Inserts date to existing series, or create new series if not already exists
• Update statistics : Number of observations, first and last observations, updated time etc
• Delete loadsets
• Delete Data on a timeseries
• Delete a timeseries and its data

• Store time series with different frequencies (weekly, monthly, quarterly, annual)
• Seamless comparison of datasets with different base years
  (e.g. 2015=100 vs 2020=100)
• Use of datetime formats enables zoomable and scrollable charts

• Aggregations:
  • avg_quarter
  • avg_year
• Functions:
  • diff_n
  • pct_1, pct_2, … pct_n (lag-based calculations)

• Chart a quarterly index together with a monthly index
• Correctly add weekly, monthly, quarterly, and annual series
• Use a common base year or base period, even if datasets differ

• Chart
• MultiChart
• YearChart
  (available under the Chart menu)

• User Report button for quick:
  • Horizontal reports
  • Vertical reports
• Report tables available directly inside charts

• Users can save and retrieve series selections
• Saved selections are automatically repopulated with fresh data (if available)
• Supports both:
  • Dynamic intervals (e.g. always current year)
  • Static intervals (e.g. always showing data up to 2025)

• Templates can be modified (optional)
• Customize:
  • Fonts
  • Colors
  • Legends
  • Tooltips
• Modify Harmonize.css to match your design standards

• Uses Highcharts by default
• Can be replaced with other compatible JavaScript charting libraries
  (not recommended)

• The C# application overwrites MyData.js each time charts are generated
  • Copy it elsewhere if you need to preserve the data
• You may:
  • Copy and modify Chart.html templates
  • Create your own HTML templates
  • Fully customize all HTML files

• Only use intervals containing data up to one year
  • Examples: CurrentYear-1, CurrentYear-2
• Additional intervals can be added in the database

Harmonize supports high-frequency data (e.g. hourly) and correctly handles summer and winter time changes.

• Data must be stored in UTC
• Each dataset can be configured (in the Loadset table) to display in a different time zone

• Time zones affect not only visualization, but also aggregation logic
• The system converts time zones before applying functions and aggregations

• Hourly spot prices:
  • 25 observations when switching to winter time
  • 23 observations when switching to summer time
• Charts may show two observations at 02:00 once per year when CET clocks are set back

• The user interface supports multiple languages
• Users can:
  • Change translations themselves
  • Add new languages as needed

Harmonize provides a powerful yet user-friendly platform for:

• Managing time series data
• Comparing datasets across frequencies and base years
• Generating interactive charts and reports
• Integrating with Python and R for advanced analytics

It is suitable for analysts, managers, and non-programmers alike.