Inflation · United States
United States inflation forecast
CPI inflation (12-month change, monthly)
Model run: Oct 14, 2026 · TimesFM 2.5
3 months forecast · Dec 2026
3.9%
Median forecast · p10-p90: 2.7% to 5.1%
6 months forecast · Mar 2027
3.7%
Median forecast · p10-p90: 2.0% to 5.5%
12 months forecast · Sep 2027
3.4%
Median forecast · p10-p90: 1.2% to 5.9%
Latest observed: 3.7% · Sep 2026
Choose a horizon
The median and range update together.
History & forecast
White: observed history. Blue: forecast median. Shaded: p10-p90 model range.
All values in 12-month change, monthly
CPI-U monthly 12-month change. October 2025 has no official observation; the gap is preserved.
Model validation
| Horizon | Tests | Model errorModel | No-change errorNo change | Band coverageCoverage |
|---|---|---|---|---|
| 3 months | 57 | 0.67 pp | 0.73 pp | 91.2% |
| 6 months | 54 | 1.01 pp | 1.08 pp | 88.9% |
| 12 months | 49 | 1.32 pp | 1.67 pp | 91.8% |
In these tests the model had lower average error than a no-change forecast at all 3 horizons. Rolling backtests of this model on the latest revised history; overlapping forecast errors are not independent. The p10-p90 band is a model range, not a calibrated guarantee. Methodology · Forecast audit
Evidence & context
Source data
Source & observed data
| Date | Value | Change |
|---|---|---|
| Sep 2026 | 3.7% | +0.34 pp |
| Aug 2026 | 3.4% | +0.03 pp |
| Jul 2026 | 3.4% | −0.17 pp |
| Jun 2026 | 3.5% | −0.72 pp |
| May 2026 | 4.2% | +0.44 pp |
| Apr 2026 | 3.8% | +0.55 pp |
| Mar 2026 | 3.3% | +0.84 pp |
| Feb 2026 | 2.4% | +0.03 pp |
| Jan 2026 | 2.4% | −0.29 pp |
| Dec 2025 | 2.7% | −0.06 pp |
| Nov 2025 | 2.7% | −0.28 pp |
| Sep 2025 | 3.0% | +0.10 pp |
Sources & limits
Source: U.S. Bureau of Labor Statistics · CUUR0000SA0. Methodology · Forecast audit
Historical context
Outside the model
What could change this forecast?
A small, dated reading list. TimesFM sees the data series, not these stories. This context is not a trading signal.
No recent item selected for this series. We would rather leave this quiet than fill it with a weak match.
Selected for relevance, not a claim of causation. Source dates stay visible. Original reporting remains with its publisher.