AgenticInvestor

Austrian CEGH VTP Natural Gas Futures

Energy · Monthly series · CM-AVM1.ICEENDEX

Latest, October 2026

105.66 EUR/MWH

+43.0% from September 2026

Open interactive chartAustrian CEGH VTP Natural Gas Futures in the app

312.1158.75.276April 2019October 2027
History through October 2026, the forecast path to October 2027 (dashed) and its 80% interval (shaded).

Forecast

Forecast for October 2027: 77.11 EUR/MWH, with an 80% interval of 5.81 EUR/MWH to 312.06 EUR/MWH.

Forecast path. Intervals are shown where available.
PeriodForecast80% low80% high
November 2026109.25 EUR/MWH76.76 EUR/MWH151.62 EUR/MWH
December 2026107.75 EUR/MWH55.09 EUR/MWH189.52 EUR/MWH
January 2027105.52 EUR/MWH43.12 EUR/MWH213.92 EUR/MWH
February 2027100.83 EUR/MWH68.31 EUR/MWH165.84 EUR/MWH
March 202799.09 EUR/MWH37.22 EUR/MWH229.75 EUR/MWH
April 202794.43 EUR/MWH30.15 EUR/MWH248.45 EUR/MWH
May 202790.26 EUR/MWH38.82 EUR/MWH223.41 EUR/MWH
June 202786.86 EUR/MWH38.79 EUR/MWH218.17 EUR/MWH
July 202784.69 EUR/MWH27.67 EUR/MWH251.71 EUR/MWH
August 202781.30 EUR/MWH12.85 EUR/MWH292.62 EUR/MWH
September 202778.83 EUR/MWH16.09 EUR/MWH282.87 EUR/MWH
October 202777.11 EUR/MWH5.81 EUR/MWH312.06 EUR/MWH

Model: Agentic Investor TSFM (foundation model). Forecast generated 2026-11-14.

AI confidence

Limited evidence

At least 6 complete tests are required; 2 are available.

Model: Agentic Investor TSFM (foundation model)

Accuracy vs naive
n/aToo few tests (2 tests, 6 needed) (2 tests, 1 won)
Direction
n/aToo few tests (2 tests, 6 needed) (15 of 24 calls right in 2 tests)
Turning points
n/aToo few tests (2 tests, 6 needed) (13 turns in 2 tests, 0 without a turn)
Error vs baseline
19% higherError ratio 1.19; 1.00 is a tie.
Tests won
1 / 2Complete recent tests excluded from model selection; ties count as no win.
Test period
2024-11-01 to 2026-10-0112 observations ahead per test.
80% band coverage
83%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

Agentic Investor TSFM scored lowest on the 2 older tests (0.87) and was chosen. The 2 recent tests played no part in the choice and give its grade (1.19). The search for a model that beats the naive forecast on both sets of tests goes on. Scores are MASE: a model's errors divided by the naive forecast's, so 1 is a tie.

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

Every model tested on this series.
ModelClass and computeOlder testsRecent testsRole
Agentic Investor TSFM (foundation model)Foundation model · CPU0.871.19Serving
Naive forecast (last value held)Benchmark · CPU1.001.00Benchmark
Exponential smoothing (statistical)Statistical · CPU1.001.00Tested
State-space model, 167M (foundation model)Foundation model · CPU1.001.02Tested
Gradient-boosted trees (machine learning)Machine learning · CPU2.051.03Tested
Hierarchical neural network (neural network)Neural network · CPU3.351.38Tested

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 92% of the 90 earlier readings since April 2019.

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
October 2026105.66 EUR/MWH
September 202673.88 EUR/MWH
August 202670.99 EUR/MWH
July 202659.94 EUR/MWH
June 202644.89 EUR/MWH
May 202647.58 EUR/MWH
April 202647.57 EUR/MWH
March 202654.36 EUR/MWH
February 202634.40 EUR/MWH
January 202641.34 EUR/MWH
December 202531.05 EUR/MWH
November 202531.63 EUR/MWH
October 202533.43 EUR/MWH
September 202532.97 EUR/MWH
August 202534.32 EUR/MWH
July 202538.56 EUR/MWH
June 202536.26 EUR/MWH
May 202537.17 EUR/MWH
April 202534.68 EUR/MWH
March 202542.34 EUR/MWH
February 202545.71 EUR/MWH
January 202554.03 EUR/MWH
December 202450.29 EUR/MWH
November 202448.59 EUR/MWH

Source

Data: public market data feeds.

More from Agentic Investor

Also stored for Energy