Semiconductor signals are split. Verify each stock separately.
0 of 2 stocks lead their benchmark over 20 days; 1 show foreign buying. The verdict stays conditional until flow and price agree in the same name.
What the engine measured
SK Hynix — 20-day relative strength is -10.4pp; the latest measured foreign flow is selling or unavailable.
4 linked sources
Samsung Electronics — 20-day relative strength is -3.1pp; the latest measured foreign flow is buying.
6 linked sources
What challenged the verdict
SK Hynix's 4.82x forward P/E depends on forecast earnings and cannot confirm undervaluation by itself.
1 linked sources
Samsung Electronics's 5.13x forward P/E depends on forecast earnings and cannot confirm undervaluation by itself.
1 linked sources
When this verdict changesAn observable condition that retires the current thesis when it occurs.
SK Hynix
Update the verdict if foreign flow and 20-day relative strength move together against the current SK Hynix assessment.
Samsung Electronics
Update the verdict if foreign flow and 20-day relative strength move together against the current Samsung Electronics assessment.
Audit method
Does upside probability multiplied by payoff still exceed the downside after the move is priced in?
What observable trigger would prove this verdict wrong? Without one, it is a story, not a testable claim.
Is this probability measured and calibrated, or is it a narrative estimate that requires smaller sizing?