| { |
| "sample_id": "0077e979b13f48c0cb8c6c50", |
| "geography": { |
| "admin1": "Hadramawt", |
| "admin2": "Hajr", |
| "country": "Yemen", |
| "iso3": "YEM" |
| }, |
| "cutoff": "2020-07-31T23:59:59Z", |
| "last_available_phase": { |
| "availability_time": "2020-07-31T23:59:59Z", |
| "event_month": "2020-06", |
| "phase": 2 |
| }, |
| "target": { |
| "deteriorated": true, |
| "label_availability_time": "2020-11-30T23:59:59Z", |
| "phase": 3, |
| "phase_change": 1, |
| "sample_id": "0077e979b13f48c0cb8c6c50", |
| "split": "train", |
| "target_month": "2020-10" |
| }, |
| "annotation": { |
| "sample_id": "0077e979b13f48c0cb8c6c50", |
| "dataset_version": "8142c89de89cb862117d6a814be515b9f00312bb960e33cf7e994a99ed124bd3", |
| "purpose": "training_supervision_only", |
| "request_sha256": "b019386cb1bc3351d00ac9d10ec0ccbddc5fd2984c962d5cafaf872c16c4aec4", |
| "model": "gpt-5.6-terra", |
| "request_id": "chatcmpl-ENP4yKC49dZ6Zwy1vUJAQgwG81ulQ", |
| "usage": { |
| "completion_tokens": 1149, |
| "prompt_tokens": 3764, |
| "total_tokens": 4913, |
| "completion_tokens_details": { |
| "accepted_prediction_tokens": 0, |
| "audio_tokens": 0, |
| "reasoning_tokens": 0, |
| "rejected_prediction_tokens": 0, |
| "text_tokens": null |
| }, |
| "prompt_tokens_details": { |
| "audio_tokens": 0, |
| "cache_write_tokens": 3761, |
| "cached_tokens": 0, |
| "image_tokens": null, |
| "text_tokens": null |
| } |
| }, |
| "annotation": { |
| "observed_future_phase": 3, |
| "precursor_patterns": [ |
| { |
| "provenance": "observed", |
| "statement": "The last available district IPC assessment was Phase 2 in 2020-06 (one month old at the July cutoff); the rolling series also retained Phase 2 throughout January-June, with the prior assessment dated 2019-10 through February and a 2020-02 assessment thereafter.", |
| "channels": [ |
| "ipc_last_known_phase", |
| "ipc_assessment_age", |
| "last_available_phase" |
| ], |
| "event_months": [ |
| "2019-10", |
| "2020-02", |
| "2020-06" |
| ], |
| "source_url": null, |
| "uncertainty": "The monthly rolling values are availability records, not separate monthly IPC assessments; they do not themselves show a change before the later observed Phase 3." |
| }, |
| { |
| "provenance": "observed", |
| "statement": "Country-level conflict event counts rose from 787 in January to 945-962 in March-April, while fatalities peaked at 2,115 in April. The trailing three-month event total remained high at 2,711-2,732 in April-May before 2,616 in June.", |
| "channels": [ |
| "acled_events", |
| "acled_fatalities", |
| "acled_events_sum_3m" |
| ], |
| "event_months": [ |
| "2020-03", |
| "2020-04", |
| "2020-05", |
| "2020-06" |
| ], |
| "source_url": null, |
| "uncertainty": "These are national counts and cannot establish exposure, displacement, market disruption, or causal effects in Hajr." |
| }, |
| { |
| "provenance": "observed", |
| "statement": "The national imported-rice price was flat at USD 3.61/kg through April, then increased 3.05% to USD 3.72/kg in May and remained there in June.", |
| "channels": [ |
| "wfp_staple_price", |
| "wfp_staple_price_change_1m_pct", |
| "wfp_staple_price_change_3m_pct" |
| ], |
| "event_months": [ |
| "2020-05", |
| "2020-06" |
| ], |
| "source_url": null, |
| "uncertainty": "This is a country-level price series for one imported staple; local prices, purchasing power, and household access in Hajr are unobserved." |
| }, |
| { |
| "provenance": "observed", |
| "statement": "National cargo imports declined relative to the prior three-month mean in April (-6.65%) and May (-12.50%), then rose in June (+27.35%); June cargo and dry-bulk import tonnage were the highest values in the six-month window. Cargo and dry-bulk measures are not food-specific.", |
| "channels": [ |
| "portwatch_import_cargo", |
| "portwatch_import_dry_bulk", |
| "portwatch_import_cargo_vs_previous_3m_pct" |
| ], |
| "event_months": [ |
| "2020-04", |
| "2020-05", |
| "2020-06" |
| ], |
| "source_url": null, |
| "uncertainty": "National port flows do not measure food availability, distribution, or access in this district, and the June rebound makes the directional signal mixed." |
| }, |
| { |
| "provenance": "observed", |
| "statement": "District normalized FCS rose by 0.074 in June after declines in April-May, ending 0.013 above March; normalized rCSI changed little in June (+0.003) and was 0.043 lower than in March. The severity direction of these normalized measures is undocumented.", |
| "channels": [ |
| "fcs_rt mean", |
| "fcs_rt mean_change_1m", |
| "fcs_rt mean_change_3m", |
| "rcsi_rt mean", |
| "rcsi_rt mean_change_1m", |
| "rcsi_rt mean_change_3m" |
| ], |
| "event_months": [ |
| "2020-03", |
| "2020-04", |
| "2020-05", |
| "2020-06" |
| ], |
| "source_url": null, |
| "uncertainty": "Numeric index movements cannot be interpreted as changes in food consumption or coping behavior without the normalization method and direction." |
| }, |
| { |
| "provenance": "inference", |
| "statement": "The subsequent IPC Phase 2-to-3 deterioration by 2020-10 occurred after a period combining elevated national conflict activity and a modest national imported-rice price increase, despite mixed import and normalized-index patterns. These channels are plausible concurrent stressors but do not demonstrate the district-level mechanism of deterioration.", |
| "channels": [ |
| "acled_events", |
| "acled_fatalities", |
| "wfp_staple_price", |
| "portwatch_import_cargo", |
| "fcs_rt mean", |
| "rcsi_rt mean", |
| "last_available_phase" |
| ], |
| "event_months": [ |
| "2020-04", |
| "2020-05", |
| "2020-06", |
| "2020-10" |
| ], |
| "source_url": null, |
| "uncertainty": "Temporal association is not causation; key local drivers such as income, displacement, assistance, market functioning, and household food access are absent." |
| } |
| ], |
| "rationale": "Observed outcome: Hajr deteriorated from the latest available IPC Phase 2 assessment in June 2020 to Phase 3 in October 2020. Before the outcome, measured national signals show conflict activity peaking in March-April and imported-rice prices rising in May, while imports first fell relative to their recent average and then rebounded strongly in June. District HFID normalized FCS and rCSI values fluctuated numerically, but their undocumented normalization precludes a food-security interpretation. The evidence therefore supports only a cautious retrospective association between broader stress signals and the later IPC deterioration.", |
| "uncertainty": "The June IPC assessment is the actual latest pre-cutoff assessment and was one month old at cutoff. Most potential precursors are national rather than Hajr-specific, rainfall is missing for June, and the available indicators omit local purchasing power, livelihoods, aid, displacement, and market conditions. No causal attribution is warranted.", |
| "confidence": 0.66 |
| } |
| } |
| } |
|
|
