Taiwan Fiscal Expenditure

The point forecast beat the no-change baseline overall and in a majority of the evaluated tests.

Model: Agentic Investor TSFM (foundation model)

Accuracy vs naive

+12%Beats the no-change baseline over 10 tests (10 tests, 7 won)

Direction

80%No clear edge over a coin flip (24 of 30 calls right in 10 tests)

Turning points

1 of 9Anticipated some turns (9 turns in 10 tests, 3 without a turn)

Error vs baseline

12% lowerError ratio 0.88; 1.00 is a tie.

Tests won

7 / 10Complete rolling backtests; ties count as no win.

Test period

1996-12-01 to 2025-12-013 observations ahead per test.

80% band coverage

80%Observed outcomes inside the band across all stored test windows.

• Band coverage uses all stored test windows, including model-selection windows when present; it does not determine the confidence label.
• Evaluated 2026-11-15 against the cached history through 2025-12-01.
• This is an evaluation of the serving model; it does not establish a winner across four model classes.