GAMMAINV Function (Gnumeric)

• Introduced: Gnumeric 0.25, explicitly named in its release notes.
• Supported in: Gnumeric 1.12.62, confirmed by its function registration. No retirement is established.
• Scope: Desktop Gnumeric with the Statistics Functions plugin. Tables use English comma-separated formulas.
• Checked: October 14, 2026. Tables were recalculated in 1.12.57 and finite reference values independently checked. Version 1.12.62 was source-checked, not executed.

GAMMAINV returns a measurement-scale value x whose gamma left-tail probability is p. Alpha is shape and beta is scale. The output uses beta’s units; p has no units. The median usually differs from the mean αβ, so entering p=0.5 does not simply multiply the two parameters. Official function reference.

=GAMMAINV(p,alpha,beta)

All arguments shown are required. The version-pinned help and evaluator establish the Gnumeric contract.

The table records the tested release’s outputs, with finite mathematical references rounded to twelve decimal places. Keep full precision when using a result in a later calculation.

For shape 2 and scale 3, enter 0.95, 2 and 3 in A2:C2. =GAMMAINV(A2,B2,C2) gives approximately 14.231593555172. Under that model, 95% of the distribution lies below the cutoff. The mean is 6, so this upper percentile is not a mean estimate. Use a proportion such as 95% or 0.95 rather than the number 95.

If a gamma model is specified by mean 6 and SD 3, calculate shape =(6/3)^2=4 and scale =3^2/6=1.5. =GAMMAINV(0.95,4,1.5) is approximately 11.630484791899. These moment relationships choose gamma parameters; they do not prove that a gamma distribution fits the observed data.

A useful scale check is GAMMAINV(p,alpha,2*beta)=2*GAMMAINV(p,alpha,beta). For shape 1, the quantile is also =-beta*LN(1-p), an independent exponential-model calculation. For ordinary interior probabilities, =GAMMADIST(GAMMAINV(p,alpha,beta),alpha,beta,TRUE()) should recover p when the same parameters and unrounded values are used.

Probability zero gives zero. Probability one reaches an unbounded mathematical quantile and produced #NUM! in the tested release. Nonpositive shape or scale and probabilities outside [0,1] are invalid. A very large or very small valid probability can require precision checks beyond these ordinary examples. Keep the underlying quantile at full precision if it feeds another probability calculation. Nonfinite numerical results are converted to worksheet errors.

These additional outcomes were observed in the same native recalculation. F5 was genuinely empty; it is not an omitted argument. Missing inputs still need deliberate validation, even if they are converted to a numeric zero.

The worked calculations do not certify every locale, imported text value, extreme parameter or later binary version. Shared numerical helpers can change even when a function’s wrapper remains the same.

• Gnumeric function reference
• Gnumeric 1.12.62 evaluator

The reference and boundary tables use recorded 1.12.57 command-line recalculation. Further model examples were independently calculated from their stated inputs; they are not additional native test claims.