Notebook environment

Managed JupyterLab on sovereign compute, governed datasets pre-mounted, TASMU SDK preinstalled.

Read-only session snapshot

q3_labour_exploration.ipynbKernel: tasmu-py311 · idle

Labour demand: Q3 2026 exploration

Exploring the latest LFS wave before the quarterly forecast release. Data mounted read-only from npc/labour-force-survey-microdata (entitlement #ENT-2214).

[1]:

from tasmu import datasets

lfs = datasets.load("npc/labour-force-survey-microdata", wave="2026Q2")
lfs.shape

(48212, 37)

[2]:

lfs.groupby("sector")["employed"].sum().nlargest(5)

Construction	412,806
Wholesale & Retail	198,344
Manufacturing	121,562
Transport & Logistics	98,417
Hospitality	91,203

[3]:

# Work-permit inflow vs LFS demand signal, drift check
from tasmu import monitoring
monitoring.psi(lfs["permit_inflow"], reference="2026Q1")

PSI = 0.31, moderate shift detected (threshold 0.25)

Note for the team: the permit-inflow feature has shifted materially since Q1, consistent with the drift advisory on labour-demand-forecaster. Recommend triggering the no-code retrain before the Q3 release.