PCE Gauge the Fed's index, re-priced daily under BEA shares
The Fed targets PCE, not CPI. This variant keeps every live component of the gauge and swaps the CPI relative-importance weights for hand-seeded BEA expenditure shares — less shelter, far more medical care — then grades itself against the official PCE price index (PCEPI).
Cite
MacroGauge PCE-weighted gauge YoY, 2026-10-02, 2018-01=100, 3.2% YoY — https://macrogauge.vercel.app/pcePCE gauge vs official PCEPI — monthly, since 2019
Export data
Next PCE print — nowcast
| Forecaster | MoM | Type | As of |
|---|---|---|---|
| Macrogauge | 0.35% | MODEL | 2026-10-02 |
| Cleveland | 0.43% | BENCHMARK | 2026-09-30 |
Core PCE (PCE less food & energy):
| Forecaster | MoM | Type | As of |
|---|---|---|---|
| Macrogauge | 0.19% | MODEL | 2026-10-02 |
| Cleveland | 0.25% | BENCHMARK | 2026-09-30 |
The PCE nowcast targets the next BEA Personal Income & Outlays release. Between the CPI and PCE prints (about two weeks) it bridges from the published seasonally adjusted CPI for that month; before the CPI print, from our CPI nowcast. Once the month's PPIs are out, the bridge adds the PPIs BEA prices PCE items with (airfares, physician and hospital care, portfolio management) — only while that bridge beats the CPI-only one out of sample. Every coefficient is published in nowcast_latest.json.
PCE calls — graded against every print
| Badge | Forecast MoM | Actual MoM | Error | Called on | Graded on | |
|---|---|---|---|---|---|---|
| 2026-08 | LIVE | 0.34% | 0.31% | +0.03pp | 2026-09-29 | 2026-09-30 |
| 2026-07 | LIVE | 0.11% | 0.16% | −0.05pp | 2026-08-12 | 2026-08-26 |
| 2026-06 | LIVE | 0.08% | -0.11% | +0.19pp | 2026-07-14 | 2026-07-30 |
| 2026-09 | pending | 0.35% | — | — | 2026-10-02 | — |
Core PCE (PCEPILFE) calls:
| Badge | Forecast MoM | Actual MoM | Error | Called on | Graded on | |
|---|---|---|---|---|---|---|
| 2026-08 | LIVE | 0.27% | 0.25% | +0.02pp | 2026-09-29 | 2026-09-30 |
| 2026-09 | pending | 0.19% | — | — | 2026-10-02 | — |
The monthly PCE nowcast is a CPI (plus PPI) pass-through; the daily PCE gauge above is a separate object — a re-weighting of live prices, not a forecast.
Weights — CPI relative importance vs BEA PCE share
| Component | CPI weight | PCE weight | Shift |
|---|---|---|---|
| Everything else (CPI residual) | 17.7% | 28.5% | +10.77pp |
| Medical care | 8.4% | 17.0% | +8.58pp |
| Shelter (owned) | 26.2% | 10.5% | −15.70pp |
| Recreation | 5.1% | 8.5% | +3.36pp |
| Food at home | 8.3% | 7.5% | −0.83pp |
| Food away from home | 5.4% | 6.5% | +1.13pp |
| Rent | 7.8% | 5.0% | −2.84pp |
| Education & comm | 5.8% | 5.0% | −0.85pp |
| Apparel | 2.4% | 2.8% | +0.43pp |
| New vehicles | 3.8% | 2.5% | −1.34pp |
| Gasoline | 2.9% | 2.2% | −0.70pp |
| Electricity | 2.5% | 2.0% | −0.49pp |
| Used cars & trucks | 2.8% | 1.2% | −1.56pp |
| Piped gas | 0.8% | 0.8% | +0.03pp |
PCE shares are hand-seeded approximations of BEA's expenditure weights over our 14 coarse components (config/basket.json, pce_weight), not a BEA-published table. They sum to one; the biggest swing is medical care (PCE counts employer- and government-paid care that CPI excludes) against shelter.