Skip to content
WorkResearchAboutContactGitHub
All work

05Climate Macro

ENSO Macro Risk Desk

When the ENSO cycle shifts, which commodity exposures are causally real, and which are spurious?

A live commodity risk desk built around the El Niño–Southern Oscillation: Niño-3.4 ONI tracking with a SARIMA+LSTM forecast cone, country-level exposure indices, and, the core idea, Granger + CCM causal testing of every headline ENSO→commodity trade.

Open the dashboard
Domain
Climate Macro
Artefact
Dashboard
Basis
causal-tests 6 headline ENSO trades
Stack
Python · Granger / CCM · SARIMA + LSTM · NOAA CPC data · Docker

What it produced

6
headline links causal-tested
11
exposed regions ranked
90%
forecast cone band (SARIMA+LSTM)

0.32

0strongest of six tested links1

Convergent cross-mapping strength for the strongest of the six headline ENSO to commodity-price links. Cross-map skill runs from 0 to 1 by construction, and none of the six clears the bar for a strong causal reading, so the desk fails its own headline trades rather than repeating them.

Source: Highlight 1 above: of six headline links, none test strongly causal, max CCM rho 0.32

ENSO Macro Risk Desk, application screenshot

The problem

Every El Niño advisory triggers the same headline trades: long palm oil, long cocoa, short wheat. Most of those links never survive causal testing; desks reposition on correlations that are confounded or spurious.

The approach

NOAA CPC ONI data feeds a live desk: a 24-month ONI trajectory with a 12-month ensemble forecast cone (SARIMA+LSTM, 90% band), an ENSO exposure index across producing countries, and an in-repo Granger + CCM engine that stress-tests each ONI→commodity-price link on linearly-detrended series. A misattribution guard flags which trades are MODERATE, WEAK, or confounded.

What it found

  1. Causal verdict: of six headline ENSO→price links, none test strongly causal (max CCM ρ 0.32)

  2. Live Niño-3.4 ONI gauge with NOAA CPC advisory status and 12-month forecast cone

  3. Exposure leaderboard: Indonesia palm oil, Côte d'Ivoire cocoa, Australia wheat and more

  4. Region deep-dives where the clean ENSO signal actually lives: climate & production data