Skip to content

Revise the IV score to base volatility on the change in N_d - #1316

Merged
WesIngwersen merged 4 commits into
mainfrom
iv-score-on-nd
Nov 18, 2026
Merged

WesIngwersen merged 4 commits into
mainfrom
iv-score-on-nd

Conversation

@WesIngwersen

@WesIngwersen WesIngwersen commented Nov 17, 2026 •

Copy link
Copy Markdown
Member

Revises the IV score so that volatility is based on the change in N_d, not the change in B.

What changes

  • The headline score is per N_d. IV_Nd_k = w_k * V_k * 100, where V_k is the existing volatility measure applied to each commodity's real N_d. The summed score is the weighted average movement of N_d, in percentage points a year.
  • Each N_d move is split by source. Each year's change in N_d splits exactly into a part from each B and a part from L_dom. IV_Nd.csv reports, for each N_d:
    • the share of its movement from B versus L_dom
    • the share from its own B
    • the single B that moves it most
  • The B score stays as the list of emission factors to investigate. IV_score.csv is unchanged, so the deep-dive triage reads it as before.
  • About_IV_score.md and the folder README are updated to describe both scores.

Results

The results for the top 50 under spend weights are in an HTML page: N_d IV Scores. It has a sortable table, and for each commodity, line charts of N_d, L_dom, B and E. All results use GHG FBS and MUT v1.0.0_2b1164f.

  • Under spend weights the summed score is 10.44. That compares with 7.76 for the B score.
  • About two-thirds of the movement comes from L_dom. Across all commodities, 68% of the summed score comes from L_dom and 32% from B.
  • Services lead the list. Software publishers, advertising and surgical instruments are at the top, and most of their movement comes from L_dom. Where B drives the movement, electricity generation's B moves them most.

Checks

--check adds two assertions, and both pass:

  • B L_dom equals N_total on the real basis.
  • The B and L_dom parts add to each annual N_d change, with a largest gap of 3.3e-15.

This PR changes only docs and an analysis script, so no CI gates apply.

🤖 Generated with Claude Code

WesIngwersen and others added 3 commits November 17, 2026 16:41
…m parts

The headline ranks each domestic total factor N_d on its own movement,
weighted by how much it counts, and names the B or L_dom part behind it.
The B score stays as the list of emission factors to investigate.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Splits each score into its B and L_dom parts and names the B that moves
each N_d most.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…an artifact

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@WesIngwersen WesIngwersen changed the title Score the IV on N_d volatility, split into B and L_dom parts Revise the IV score to base volatility on the change in N_d Nov 17, 2026
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

@bl-young bl-young left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good. Scoring headline volatility on N_d (with the exact B / L_dom split) addresses the real gap in the B-only IV score, and keeping IV_score.csv unchanged preserves the triage path. Docs match the dual-score framing.

@WesIngwersen
WesIngwersen merged commit b7e98ac into main Nov 18, 2026
5 checks passed
@WesIngwersen
WesIngwersen deleted the iv-score-on-nd branch November 18, 2026 02:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants