Japan Construction Output YoY

Japan · Monthly series · EC-JP-COUT

Latest, July 2020

-6.9%

+1.00pp from June 2020

Open interactive chartJapan Construction Output YoY in the app

14.5-0.4293-15.36August 2010July 2021
History through July 2020, the forecast path to July 2021 (dashed) and its 80% interval (shaded).

Forecast

Forecast for July 2021: -3.1%, with an 80% interval of -15.4% to 9.4%.

Forecast path. Intervals are shown where available.
PeriodForecast80% low80% high
August 2020-6.2%-7.6%-4.9%
September 2020-6.2%-9.1%-2.8%
October 2020-6.1%-8.8%-3.1%
November 2020-5.8%-9.5%-1.5%
December 2020-5.7%-10.3%-0.3%
January 2021-5.5%-8.5%-1.9%
February 2021-5.2%-8.7%-1.0%
March 2021-4.6%-9.6%1.0%
April 2021-4.2%-13.7%6.8%
May 2021-3.7%-14.4%7.5%
June 2021-3.4%-14.0%8.4%
July 2021-3.1%-15.4%9.4%

Model: Encoder-only transformer, 120M (foundation model). Forecast generated 2026-11-14.

AI confidence

Needs refresh

The data refresh is due. Confidence will be reassessed against the refreshed history.

Model: Encoder-only transformer, 120M (foundation model)

Accuracy vs naive
n/aNot evaluated
Direction
n/aNot evaluated
Turning points
n/aNot evaluated
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.

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 2% of the 119 earlier readings since August 2010.

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
July 2020-6.9%
June 2020-7.9%
May 2020-6.4%
April 2020-3.1%
March 2020-3.0%
February 2020-4.6%
January 2020-2.5%
December 2019-1.7%
November 2019-1.2%
October 2019-0.9%
September 2019-1.4%
August 20190.2%
July 20190.9%
June 20191.4%
May 20190.9%
April 2019-0.8%
March 2019-1.4%
February 2019-2.1%
January 2019-3.9%
December 2018-5.6%
November 2018-3.1%
October 2018-3.7%
September 2018-2.7%
August 2018-2.6%

Source

Data: public market data feeds.

More from Agentic Investor

Also stored for Japan