Taiwan Tourist Arrivals

Taiwan · Monthly series · EC-TW-TA

Latest, September 2026

611,760.0 pts

-7.7% from August 2026

Open interactive chartTaiwan Tourist Arrivals in the app

1.143M572.9K2559October 2016September 2027
History through September 2026, the forecast path to September 2027 (dashed) and its 80% interval (shaded).

Forecast

Forecast for September 2027: 627,689.2 pts, with an 80% interval of 420,897.0 pts to 799,338.7 pts.

Forecast path. Intervals are shown where available.
PeriodForecast80% low80% high
October 2026709,721.9 pts622,345.9 pts809,415.9 pts
November 2026780,414.1 pts666,506.6 pts893,004.4 pts
December 2026888,589.2 pts645,594.0 pts1,097,501.4 pts
January 2027681,088.4 pts533,089.4 pts841,700.1 pts
February 2027690,924.4 pts526,447.5 pts861,020.1 pts
March 2027820,282.2 pts531,878.6 pts1,057,066.5 pts
April 2027751,911.9 pts514,685.7 pts956,500.2 pts
May 2027695,036.4 pts538,281.9 pts838,206.8 pts
June 2027629,066.5 pts469,076.7 pts777,336.7 pts
July 2027616,347.4 pts419,608.4 pts777,484.5 pts
August 2027671,803.9 pts462,591.1 pts841,553.8 pts
September 2027627,689.2 pts420,897.0 pts799,338.7 pts

Model: Agentic Investor TSFM (foundation model). Forecast generated 2026-11-15.

AI confidence

Supported evidence

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
+17%Beats the no-change baseline over 12 tests (12 tests, 10 won)
Direction
79%Beats a coin flip (114 of 144 calls right in 12 tests)
Turning points
53 of 59Anticipated most turns (59 turns in 12 tests, 0 without a turn)
Error vs baseline
17% lowerError ratio 0.83; 1.00 is a tie.
Tests won
10 / 12Complete rolling backtests; ties count as no win.
Test period
2014-10-01 to 2026-09-0112 observations ahead per test.
80% band coverage
80%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

Comparison pendingA new ticker first receives its serving model's forecast and evaluation. An eligible nightly batch can compare both candidates, subject to call budgets and retry limits.

Reading

It is above 46% of the 119 earlier readings since October 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
September 2026611,760.0 pts
August 2026662,740.0 pts
July 2026607,530.0 pts
June 2026611,150.0 pts
May 2026692,780.0 pts
April 2026758,610.0 pts
March 2026883,060.0 pts
February 2026625,610.0 pts
January 2026723,380.0 pts
December 2025946,540.0 pts
November 2025812,240.0 pts
October 2025757,860.0 pts
September 2025601,340.0 pts
August 2025673,100.0 pts
July 2025586,230.0 pts
June 2025606,350.0 pts
May 2025710,930.0 pts
April 2025751,150.0 pts
March 2025778,580.0 pts
February 2025699,160.0 pts
January 2025651,080.0 pts
December 2024903,619.0 pts
November 2024747,800.0 pts
October 2024661,115.0 pts

Source

Data: public market data feeds.

More from Agentic Investor

Also stored for Taiwan