Change to period-end data

[idelder]

While Temoa indexes periods by period-start (e.g., 2025 for 2025-2030), we now plan for END of period, with all end of period data (e.g., 2030 data for 2025-2030, though it is still labelled 2025 in the database and in Temoa).

[Copilot]

Pull request overview

This PR updates the industry ETL to align with an end-of-period data convention (e.g., model period 2025 representing 2030 “data year”), while keeping the period labels unchanged for Temoa/database compatibility.

Changes:

• Add a data_year() helper and use it to annotate notes with end-of-period “data year” context.
• Update demand GDP-year filtering to include both period years and their end-of-period years for growth calculations.
• Adjust runtime configuration (version/period list), data_id prefixing, and add a .gitignore for generated artifacts.

Reviewed changes

Copilot reviewed 6 out of 7 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
common.py	Adds data_year() helper to convert model period/vintage to representative end-of-period data year.
demands.py	Incorporates end-of-period years into GDP filtering and uses data_year() when scaling demand values/notes.
techinput.py	Updates notes to include end-of-period “data year” via data_year().
efficiency.py	Updates notes to include end-of-period “data year” via data_year().
setup.py	Changes generated data_id prefix for industry high-resolution IDs.
input/params.yaml	Updates version and the configured periods list.
.gitignore	Ignores generated outputs and Python bytecode/cache files.

Comments suppressed due to low confidence (1)

demands.py:142

• GDP scaling is computed as a ratio between consecutive years in all_gdp_years (e.g., 2030/2025), but the demand baseline you multiply by is explicitly 2022 (base_2022). This makes val = base_2022 * scale inconsistent (it should be scaled relative to the baseline year’s GDP, e.g., GDP(dy)/GDP(2022), or you should change the baseline year to match the GDP anchor). Consider building a year->GDP-level map from pop_df and computing scale = gdp_level[dy] / gdp_level[baseline_year] (and include the baseline year in the filtered GDP years).

    # ---- GDP scaling dict from CER CEF GNZ ----
    # End-of-period convention: include both period years and their end-of-period
    # data years so we can compute growth rates that span each period.
    end_years = [data_year(p, periods) for p in periods]
    all_gdp_years = sorted(set(periods) | set(end_years))

    gdp_df = pop_df.copy()
    gdp_df = gdp_df[gdp_df['Year'].isin(all_gdp_years)]
    gdp_df = gdp_df[gdp_df['Variable'] == 'Real Gross Domestic Product ($2012 Millions)']
    gdp_df = gdp_df[gdp_df['Scenario'] == 'Global Net-zero']
    gdp_df = gdp_df.sort_values('Year').reset_index(drop=True)

    # gdp_dict[y] = GDP(y) / GDP(previous year in filtered set), first entry = 1.0
    gdp_dict: dict[int, float] = {}
    for i, row in gdp_df.iterrows():
        year = int(row['Year'])
        val = float(row['Value'])
        if i == 0:
            gdp_dict[year] = 1.0
        else:
            prev_val = float(gdp_df.loc[i - 1, 'Value'])
            gdp_dict[year] = (val / prev_val) if prev_val != 0 else 1.0

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.

[Copilot]

data_year will raise an IndexError if model_periods is an empty list (e.g., if periods: [] is provided in params.yaml). It would be better to validate upfront and raise a clear ValueError (or provide a default) so failures are easier to diagnose.

[Copilot]

Typo in the inline comment: “most reason year” should be “most recent year”.

Suggested change

	NRCan_year: 2022 #Check the NRCan website and find the most reason year in the database, that is the value to use
	NRCan_year: 2022 #Check the NRCan website and find the most recent year in the database, that is the value to use