Beneish · M-Score earnings manipulation detection
An 8-variable discriminant built from two periods of financial line items, flagging abnormal changes in receivables, gross margin, asset quality, growth, depreciation, expense ratios, accruals and leverage to detect earnings-manipulation tendencies.
- Guide level
- Practical
- Output
- Screening and validation
Core formula
M = −4.84 + 0.92·DSRI + 0.528·GMI + 0.404·AQI + 0.892·SGI + 0.115·DEPI − 0.172·SGAI + 4.679·TATA − 0.327·LVGI; M > −1.78 flags suspicionHow to interpret it
An 8-variable discriminant built from two periods of financial line items, flagging abnormal changes in receivables, gross margin, asset quality, growth, depreciation, expense ratios, accruals and leverage to detect earnings-manipulation tendencies.
The output should be read as a scenario or decision aid, not as a guaranteed price target. Compare it with at least one method based on different economic assumptions.
Practical workflow
- Normalize the latest public financial and operating data.
- Choose assumptions that match the company's economics and accounting structure.
- Calculate conservative, base and optimistic cases where the method permits.
- Compare the result with market pricing and an independent valuation method.
- Document the assumptions that drive the largest changes in value.
Key limitation
Requires two full, consistent periods of financials; prone to false positives on genuinely fast-growing companies; not applicable to financials.
How Stockinsky uses it
Stockinsky uses this framework only when the company type and available data support it. Professional valuations expose assumptions, sources and warnings instead of presenting false precision.