Claim at v1.1.0
ems-traffic-hourly-elasticity
9fb108ce3cfeb784f0aee08db809c7f4539a6e08930597f585d8b6b602ade102
Data
{
"has_outcome": true,
"kind": "confirmation",
"lineage": {
"audit": "docs/lineage_audit.md; data/derived/lineage_audit.json",
"class": "c",
"class_meaning": "a shared cache: the query was fetched for an earlier claim and holds this claim's outcome",
"classes": [
"c"
],
"early_sources": [
{
"class": "c",
"dataset": "76xm-jjuj",
"earlier_claim": "bridge-crz-ems-travel",
"evidence": "The record's time (13:10:56) is the 2017 hourly calibration fetch, but the same record covers the 2019-2026 EMS rows (data/derived/congestion/ems_YYYY.csv), fetched 07:56-08:09 for bridge-crz-ems-travel and committed in 1819abd at 08:44:57, before this registration (13:27:12). Referee: those rows \"were committed ... before the registration\"; \"Nothing shows that the registrant joined them\" with taxi data before registering.",
"fetched": "2026-11-13T13:10:56-05:00",
"name": "EMS Incident Dispatch Data",
"revealed_deciding_number": "no: bridge-crz-ems-travel reported a monthly pre-period elasticity of 0.32, disclosed; the hourly within-day elasticity was not computed"
}
]
},
"outcome": "ems-traffic-hourly-elasticity",
"prereg_branch": {
"name": "prereg/ems-traffic-hourly-elasticity",
"sha": "d75fa5e45a373186fbb58bbf701634078996d8b1"
},
"prior": 0.8,
"registered_at": "2026-11-13T13:27:12-0500",
"registrant": "claude (ems_response thread)",
"registration": {
"alternatives": [
"Fewer ambulance unit-hours (MMR: in-service hours per day 8,891 in FY22 to 8,109 in FY26), so the nearest free unit is farther away (a distance effect, not traffic)",
"Hospital closures and diversions lengthening repositioning (Mount Sinai Beth Israel closed 2025)",
"Call mix and geography (more calls in slow areas or hours)",
"Recording changes (share of incidents without a valid response time rose from 4.4% in 2023 to 9.5% in 2026)",
"A shared within-hour cause, such as rain or an event, that slows traffic and also raises demand in that hour"
],
"graph_evidence": [
{
"pubmed_ids": [
"9795322",
"20798090"
],
"summary": "The graph holds no finding on the elasticity of emergency-vehicle travel time to general traffic speed. It holds a lights-and-sirens time saving of 3.02 min (Minneapolis, PMID 9795322) and survival odds ratios of 1.02 to 1.24 per minute of shorter response (docs/congestion_tradeoffs.md)."
}
],
"id": "ems-traffic-hourly-elasticity",
"kind": "confirmation",
"literature": [
{
"citation": "docs/congestion_tradeoffs.md, bridge-crz-ems-travel registration",
"finding": "Monthly zone-relative elasticity of ambulance travel time to taxi speed 0.32 (five specifications 0.14 to 0.49), October 2019 to December 2024."
},
{
"citation": "Calibration 2017-2018 (this thread, calibration.json, seen before registration)",
"finding": "Same model on 2017-2018: elasticity 0.70 (0.64 to 0.75), lead coefficient in the joint model -0.04 (-0.08 to 0.01); life-threatening only 0.53 (0.46 to 0.61)."
}
],
"plan": "lib.t1(range(2019, 2027)) at commit f44ab0b. Cells: date x hour x area (crz with taxi cbd, north with taxi north), at least 2 EMS incidents with travel > 0 and 20 taxi trips. y = log mean EMS travel seconds (76xm-jjuj, valid response time, 2019-2026 rows from data/derived/congestion/ems_YYYY.csv); x = log taxi hours per mile (TLC yellow trip files, filter f1, analyses/ems_response/taxi_hourly.py); x_lead = x of the same area and hour 7 days later. Controls log n of the cell and log Manhattan EMS n of the hour. Fixed effects date x area and area x year x weekday x hour, weights EMS n, standard errors clustered by week. No resampling. Discrimination (rule 8): the 2017-2018 calibration gives se(beta_x) = 0.030, so D against no effect at a predicted 0.4 is 181. Secondary (reported, not deciding): life-threatening only (sev hi); x alone; lead alone; by year.",
"prediction": {
"direction": "+",
"magnitude": "elasticity of mean EMS travel time to taxi pace (hours per mile) 0.3 to 0.9; lead coefficient near 0",
"refute_if": "The 95% upper bound of beta_x is below 0.10, OR the 95% upper bound of beta_x - beta_lead is below 0 (future traffic predicts at least as well as current traffic).",
"support_if": "Joint model (x and x_lead together), all severities, 2019-01-01 to 2026-09-30 without 2020-03-01 to 2021-06-30: the 95% lower bound of beta_x is above 0.10, AND the 95% lower bound of beta_x - beta_lead is above 0.10."
},
"prior": 0.8,
"registered_at": "2026-11-13T13:27:12-0500",
"registrant": "claude (ems_response thread)",
"sha256": "ad3404a9d7090050cd7910f3742ac2d8b150b6194d815eced9cf77aab7a312fa",
"statement": "Within a day, Manhattan ambulance travel time rises in hours when yellow-taxi trips in the same area are slower than usual, with an elasticity between 0.3 and 0.9, and taxi pace a week later does not predict it."
},
"registration_commit": {
"commit": "d75fa5e45a373186fbb58bbf701634078996d8b1",
"committed_at": "2026-11-13T13:27:13-05:00"
},
"registration_file_sha256": "7a233e3087417cbb578bfbea8e28dfc73eb32d059fe4a196a951d4034bd9b804",
"registration_path": "claims/registered/ems-traffic-hourly-elasticity.json",
"registration_sha256": "ad3404a9d7090050cd7910f3742ac2d8b150b6194d815eced9cf77aab7a312fa",
"registration_text": "{\n \"id\": \"ems-traffic-hourly-elasticity\",\n \"kind\": \"confirmation\",\n \"statement\": \"Within a day, Manhattan ambulance travel time rises in hours when yellow-taxi trips in the same area are slower than usual, with an elasticity between 0.3 and 0.9, and taxi pace a week later does not predict it.\",\n \"graph_evidence\": [\n {\n \"summary\": \"The graph holds no finding on the elasticity of emergency-vehicle travel time to general traffic speed. It holds a lights-and-sirens time saving of 3.02 min (Minneapolis, PMID 9795322) and survival odds ratios of 1.02 to 1.24 per minute of shorter response (docs/congestion_tradeoffs.md).\",\n \"pubmed_ids\": [\n \"9795322\",\n \"20798090\"\n ]\n }\n ],\n \"literature\": [\n {\n \"citation\": \"docs/congestion_tradeoffs.md, bridge-crz-ems-travel registration\",\n \"finding\": \"Monthly zone-relative elasticity of ambulance travel time to taxi speed 0.32 (five specifications 0.14 to 0.49), October 2019 to December 2024.\"\n },\n {\n \"citation\": \"Calibration 2017-2018 (this thread, calibration.json, seen before registration)\",\n \"finding\": \"Same model on 2017-2018: elasticity 0.70 (0.64 to 0.75), lead coefficient in the joint model -0.04 (-0.08 to 0.01); life-threatening only 0.53 (0.46 to 0.61).\"\n }\n ],\n \"prediction\": {\n \"direction\": \"+\",\n \"magnitude\": \"elasticity of mean EMS travel time to taxi pace (hours per mile) 0.3 to 0.9; lead coefficient near 0\",\n \"support_if\": \"Joint model (x and x_lead together), all severities, 2019-01-01 to 2026-09-30 without 2020-03-01 to 2021-06-30: the 95% lower bound of beta_x is above 0.10, AND the 95% lower bound of beta_x - beta_lead is above 0.10.\",\n \"refute_if\": \"The 95% upper bound of beta_x is below 0.10, OR the 95% upper bound of beta_x - beta_lead is below 0 (future traffic predicts at least as well as current traffic).\"\n },\n \"plan\": \"lib.t1(range(2019, 2027)) at commit f44ab0b. Cells: date x hour x area (crz with taxi cbd, north with taxi north), at least 2 EMS incidents with travel > 0 and 20 taxi trips. y = log mean EMS travel seconds (76xm-jjuj, valid response time, 2019-2026 rows from data/derived/congestion/ems_YYYY.csv); x = log taxi hours per mile (TLC yellow trip files, filter f1, analyses/ems_response/taxi_hourly.py); x_lead = x of the same area and hour 7 days later. Controls log n of the cell and log Manhattan EMS n of the hour. Fixed effects date x area and area x year x weekday x hour, weights EMS n, standard errors clustered by week. No resampling. Discrimination (rule 8): the 2017-2018 calibration gives se(beta_x) = 0.030, so D against no effect at a predicted 0.4 is 181. Secondary (reported, not deciding): life-threatening only (sev hi); x alone; lead alone; by year.\",\n \"alternatives\": [\n \"Fewer ambulance unit-hours (MMR: in-service hours per day 8,891 in FY22 to 8,109 in FY26), so the nearest free unit is farther away (a distance effect, not traffic)\",\n \"Hospital closures and diversions lengthening repositioning (Mount Sinai Beth Israel closed 2025)\",\n \"Call mix and geography (more calls in slow areas or hours)\",\n \"Recording changes (share of incidents without a valid response time rose from 4.4% in 2023 to 9.5% in 2026)\",\n \"A shared within-hour cause, such as rain or an event, that slows traffic and also raises demand in that hour\"\n ],\n \"prior\": 0.8,\n \"registrant\": \"claude (ems_response thread)\",\n \"registered_at\": \"2026-11-13T13:27:12-0500\",\n \"sha256\": \"ad3404a9d7090050cd7910f3742ac2d8b150b6194d815eced9cf77aab7a312fa\"\n}",
"statement": "Within a day, Manhattan ambulance travel time rises in hours when yellow-taxi trips in the same area are slower than usual, with an elasticity between 0.3 and 0.9, and taxi pace a week later does not predict it."
}History in this collection
- addedv1.0.0Nov 14, 2026this version