US
Top drivers
⌁ mcp.call("adw-314") vADW-314-live-1.0 Is US infrastructure investment accelerating or contracting relative to trend?
US
Top drivers
⌁ mcp.call("adw-314") vADW-314-live-1.0 A construction-sector investment research agent polls ADW-314 monthly and, when the score drops below 50 on a falling trend (current: 54.2, 17th percentile, down 18.3% from the 2016 start and falling), it automatically downweights infrastructure-linked equity positions in a model portfolio and appends the top_drivers field—which separates total-construction momentum from public-spend momentum—to a research note explaining whether private or government spending is the drag. Source_lineage back to Census VIP via the FRED mirror means the agent can cite the primary government data release rather than a derived vendor estimate, satisfying compliance requirements for investment rationale.
A VP of Corporate Strategy at a building-materials company uses ADW-314 to track whether the IIJA (Infrastructure Investment and Jobs Act) spending is actually showing up in construction-put-in-place figures or is still stuck in project-approval pipelines. The current 17th-percentile reading—the lowest in a decade of backtest history—tells the leadership team that velocity is contracting even against a moderate baseline, supporting a decision to defer capacity expansion capex rather than commit to a new production facility ahead of demand materializing.
0.5*total + 0.5*public; sub=clamp(50+YoY%*2.0+3m-mom%*1.5,0,100)
Version ADW-314-live-1.0 · validated to beat a naive baseline · benchmark: Dodge Construction (paid); ENR (paid)
One call returns the answer with its reasoning attached — the live Intelligence Object for ADW-314.
{
"product_id": "ADW-314",
"entity": "US",
"score": 54.2,
"trend": "stable",
"confidence": 0.74,
"top_drivers": [
{
"factor": "total_construction_yoy_pct",
"contribution": 0.88
},
{
"factor": "total_construction_mom3_pct",
"contribution": -1.24
},
{
"factor": "total_spending_latest_usd_m",
"contribution": 2172397
},
{
"factor": "public_construction_yoy_pct",
"contribution": 3.67
},
{
"factor": "public_construction_mom3_pct",
"contribution": 0.68
},
{
"factor": "public_spending_latest_usd_m",
"contribution": 532728
}
],
"recommended_use": "Gauge whether US infrastructure investment is accelerating or decelerating vs. its own trend. Score > 55 = investment is running above trend (positive tailwind for materials, construction equipment, REITs); score < 45 = spending is contracting. Use as a capex-cycle leading indicator for sector allocation. Data lags ~1-2 months (Census VIP publication cycle).",
"methodology_version": "ADW-314-live-1.0",
"freshness": "2026-06-27T03:00:27.602Z",
"coverage": "US Total Construction Spending (TTLCONS) + Total Public Construction Spending (TLPBLCONS); both from US Census Bureau Value of Construction Put in Place (C30/VIP), served via FRED; seasonally adjusted annual rate, monthly from 1993-01",
"source_lineage": [
"fred.stlouisfed.org/graph/fredgraph.csv?id=TTLCONS (keyless; canonical source = US Census Bureau VIP survey, C30 release)",
"fred.stlouisfed.org/graph/fredgraph.csv?id=TLPBLCONS (keyless; canonical source = US Census Bureau VIP survey, public-sector component)"
],
"allowed_use": "informational",
"validation_status": "descriptive",
"total_construction_sub_score": 49.9,
"public_construction_sub_score": 58.4,
"total_construction_yoy_pct": 0.88,
"total_construction_mom3_pct": -1.24,
"total_spending_latest_usd_m": 2172397,
"total_spending_latest_period": "2026-04-01",
"public_construction_yoy_pct": 3.67,
"public_construction_mom3_pct": 0.68,
"public_spending_latest_usd_m": 532728,
"public_spending_latest_period": "2026-04-01",
"n_total_periods": 400,
"n_public_periods": 400,
"score_weight_note": "50% total construction velocity, 50% public construction velocity",
"methodology_note": "sub_score = clamp(50 + YoY% × 2.0 + 3m-mom% × 1.5, 0, 100). YoY = latest month vs same month 12 months prior (Census SAAR series, seasonally adjusted, so direct comparison is valid). 3-month momentum = avg(last 3) vs avg(prior 3). Calibration: ±20% YoY shifts score ±40 pts; ±10% short-term momentum shifts ±15 pts. Final score = simple average of total and public sub-scores (equal weight). Census VIP publication lag is typically 4-6 weeks after reference month end.",
"data_lag_note": "Census Value of Construction Put in Place (C30) publishes ~4-6 weeks after the reference month. The latest data point is therefore typically 5-7 weeks old at time of fetch."
} Every product conforms to the Intelligence Object Model — typed, versioned, and discoverable.
Dashboard
Read the score + drivers in the console.
REST API
/v1/intelligence/adw-314
MCP tool
adw.adw_314
Marketplace
Discoverable by any MCP agent via the MCP registry.
White-label
Embed under your own brand (Platinum).
Is US industrial output accelerating?
Method: recent vs trailing-mean % deviation, scaled (FRED INDPRO)
Is US business capex (durable-goods orders) accelerating?
Method: recent vs trailing-mean % deviation, scaled (FRED DGORDER)