← Every exhibit

Figure Figure 2 Growth that compounds

What the stop test reports when there is nothing to report

Nothing in the simulated data depreciates. The first two designs still report a rate, and its sign follows the reason the programme was stopped rather than anything about the asset.

What five versions of the stop test report when the true rate is zero A dot-and-range chart. The horizontal axis is the annual depreciation rate the test reports, from minus 20 to plus 20 per cent, with a heavy line at zero marking the truth. Four rows. Stopping after a strong run reports a median of plus 6.8 per cent. Stopping after a weak run reports minus 6.6 per cent. Stopping on a schedule reports plus 0.8 per cent but with a range from minus 11.4 to plus 11.3. Scheduling with a matched comparison arm reports plus 0.1 per cent over a similar range. the truth: 0% Stopped after a strong run +6.8% Stopped after a weak run -6.6% Stopped on a schedule +0.8% Scheduled, with an unrelated arm +0.0% Scheduled, with a matched arm +0.1% -20-100+10+20 Annual depreciation rate the test reports (%) How the stop was chosen Dot: median of 20,000 runs. Bar: the middle 80% of runs.
Vector SVG Cite Embed

Reference & Evidence

Source: Simulation, 20,000 runs per design. An AR(1) quarterly series, persistence 0.3, standard deviation 15% of level, with a true depreciation rate of exactly zero; twelve quarters before the stop and eight after. Script: instrument-test-decay.py, filed with the claim ledger.

Related exhibits