AgenticInvestor

United States Real GDP YoY

United States · Quarterly series · EC-US-GDPYY

Latest, Q3 2026

2.1%

-0.12pp from Q2 2026

Open interactive chartUnited States Real GDP YoY in the app

12.562.58-7.4Q4 1996Q3 2027
History through Q3 2026, the forecast path to Q3 2027 (dashed) and its 80% interval (shaded).

Forecast

Forecast for Q3 2027: 1.7%, with an 80% interval of 0.5% to 2.9%.

Forecast path. Intervals are shown where available.
PeriodForecast80% low80% high
Q4 20261.8%0.8%2.7%
Q1 20271.5%-0.8%3.9%
Q2 20271.7%1.3%2.1%
Q3 20271.7%0.5%2.9%

Model: Gradient-boosted trees (machine learning). Forecast generated 2026-11-14.

AI confidence

Mixed evidence

The tests show no consistent, material advantage over the no-change baseline.

Model: Gradient-boosted trees (machine learning)

Accuracy vs naive
+41%Beats the no-change baseline over 6 tests (6 tests, 3 won)
Direction
67%No clear edge over a coin flip (16 of 24 calls right in 6 tests)
Turning points
6 of 9Anticipated most turns (9 turns in 6 tests, 0 without a turn)
Error vs baseline
41% lowerError ratio 0.59; 1.00 is a tie.
Tests won
3 / 6Complete recent tests excluded from model selection; ties count as no win.
Test period
2020-12-01 to 2026-09-014 observations ahead per test.
80% band coverage
81%Observed outcomes inside the band across all stored test windows.
How this assessment is made

AI confidence describes historical evidence for this forecast, not a probability of being correct. Each backtest fits on data available before its test window. The baseline carries the last value forward. Supported evidence requires at least six complete, non-overlapping tests, an error ratio below 0.90 (rounded to two decimals), and wins in a majority of tests. Ratios from 0.90 to below 1.10 are on par; 1.10 or above is weak evidence. Inconsistent wins produce mixed evidence. For a selected model, only the recent tests excluded from model selection determine this label. These are historical simulations using the available data, not a live trading track record; revisions and foundation-model pretraining can limit how closely they reproduce a past decision. Past performance can change.

Replay past tests

Model comparison

5 of 5 models evaluatedModels are selected on older tests and evaluated on separate recent tests. All four classes completed their backtests.

Models tested

Gradient-boosted trees scored lowest on the 2 older tests (0.93) and was chosen. The 6 recent tests played no part in the choice and give its grade (0.59). That grade beats the naive forecast. Scores are MASE: a model's errors divided by the naive forecast's, so 1 is a tie.

4 model classes completed the tests: Machine learning, Foundation model, Statistical, Neural network.

Every model tested on this series.
ModelClass and computeOlder testsRecent testsRole
Gradient-boosted trees (machine learning)Machine learning · CPU0.930.59Serving
Agentic Investor TSFM (foundation model)Foundation model · CPU0.940.61Tested
State-space model, 167M (foundation model)Foundation model · CPU0.980.66Tested
Exponential smoothing (statistical)Statistical · CPU1.001.00Tested
Naive forecast (last value held)Benchmark · CPU1.001.00Benchmark
Hierarchical neural network (neural network)Neural network · CPU1.054.15Tested

Uses ci-proto, released by Continuous Intelligence. Continuous Intelligence does not endorse or approve this product or service. (ci-proto is shown as State-space model, 167M.)

Reading

It is above 21% of the 39 earlier readings since Q4 2016.

Written from the figures by a fixed template; no AI model is involved.

Recent history (24 observations)
The 24 most recent observations, newest first.
PeriodValue
Q3 20262.1%
Q2 20262.2%
Q1 20262.6%
Q4 20252.0%
Q3 20252.6%
Q2 20252.4%
Q1 20252.3%
Q4 20242.6%
Q3 20243.0%
Q2 20243.3%
Q1 20242.9%
Q4 20233.4%
Q3 20233.2%
Q2 20232.7%
Q1 20232.3%
Q4 20221.2%
Q3 20222.2%
Q2 20222.3%
Q1 20224.0%
Q4 20215.8%
Q3 20215.3%
Q2 202112.6%
Q1 20211.9%
Q4 2020-0.9%

Source

Data: public market data feeds.

More from Agentic Investor

Also stored for United States