Taiwan Distress Distance (Altman Z) — a screen, never a verdict
When people see a distress score the question in their head is almost always “will this company fail”. This page deliberately does not answer it. Altman Z is a 1968 discriminant function fitted on US listed manufacturers; it sorts firms into three zones and never emitted a probability of failure — converting the score into a percentage is a later, baseless step. What is shown here is distance and composition: how far from the safe zone, and which component pushed you there.
What the five components ask
Working capital, retained earnings, operating profit, market equity against liabilities, and asset turnover — each returned with its raw inputs, component value and the original published coefficient. No parameter was tuned, no weight invented, no threshold added, so anyone can recompute it from the same filing.
Where we mark the model not applicable
The function was fitted on US listed manufacturers. For financial firms high leverage is the business, not a warning, and for asset-light firms the total-asset denominator loses its meaning. Those cases are marked not applicable with a reason rather than forced through — a forced number would still be displayed as “distress risk” while resting on a sample premise that does not hold.
A grey zone is reported as grey
The model sorts firms into safe, grey and distress zones. Grey is the model's own statement that it cannot discriminate in that range; it is not a polite way of saying “leaning safe”. We therefore report grey with the distance to each boundary and never nudge it either way. A per-period series is also returned, because the direction of travel and which component moved usually carries more information than one reading.
What this page is not for
The score is not a probability of failure, not a credit rating, not a trading signal. Plenty is invisible to it: off-balance-sheet commitments, related-party dealings, customer concentration, structural change. Inputs come from official filings gated by statutory publication deadlines; a missing component makes the whole reading undetermined with the gap named, never zero-filled or averaged in.