BETA.INV in Collabora Office: Find Quantiles Within Explicit Bounds
Use Collabora BETA.INV to calculate bounded-distribution quantiles, understand optional bounds, and audit inverse results with cumulative probabilities.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: =BETA.INV(0.5;2;2;10;20) has independently expected result 15. It is the median of a symmetric beta distribution scaled to the interval from ten to twenty.
Syntax and argument order
=BETA.INV(Number; Alpha; Beta [; Start [; End]])
Number is the cumulative probability. Alpha and Beta are strictly positive shape parameters. Start and End are optional output-range bounds; the dedicated Collabora help gives defaults of zero and one, respectively. Use Start<End and interior probabilities for the finite examples here.
Collabora’s help says optional arguments can be omitted only at the end. To specify a custom End, supply Start too. The formulas below deliberately show both bounds when using a nonstandard interval, avoiding an ambiguous blank placeholder.
This inverse returns the x-value associated with a cumulative probability. The help’s phrase “cumulative Beta probability density function” refers to the inverse cumulative distribution, not an inverse of the density curve. A density can have the same height at multiple positions, so that would not define the same quantile operation.
Example 1: calculate three quantiles
Put 0.1, 0.5 and 0.9 in A2:A4. Enter =BETA.INV(A2;2;2;0;1) in B2 and fill down. Independent expected results are approximately 0.195800106, 0.5 and 0.804199894.
The shape-two, shape-two distribution is symmetric about 0.5. Its tenth and ninetieth percentiles add to one. This symmetry provides a convenient check for this example, but it is not a property of every pair of beta parameters.
Omitting both bounds, =BETA.INV(0.5;2;2), also targets 0.5 using the documented defaults. Keeping explicit bounds is often clearer when a model will later be rescaled.
Example 2: verify the quantiles with a polynomial
For this particular distribution on zero to one, the cumulative function is 3*x^2-2*x^3. In C2 enter =3*B2^2-2*B2^3 and fill down. The independent targets are the original probabilities 0.1, 0.5 and 0.9.
A direct distribution check is =BETA.DIST(B2;2;2;1;0;1). The cumulative flag must be one: applying density mode to a quantile does not recover its probability. Keep the full stored B values for these checks.
If residuals are large, check whether 10 was entered instead of 10%, whether the shape arguments changed between the forward and inverse formulas, or whether a displayed quantile was manually rounded before reuse.
Advanced application: scale a quantile interval
For bounds ten and twenty, enter =BETA.INV(A2;2;2;10;20) in D2 and fill down. Independent targets are approximately 11.958001057, 15 and 18.041998943.
These should equal =10+10*B2 and its filled-down equivalents. The affine mapping is Start+(End-Start)*standard_quantile. It changes the output units while preserving each cumulative probability.
The central 80% model interval runs from the tenth to the ninetieth percentile, approximately 11.958 to 18.042 here. It is a distribution interval under the chosen model, not automatically a confidence interval for an unknown parameter or a guarantee about future observations.
Advanced application: reproduce a non-symmetric official example
=BETA.INV(0.5;5;10) has independently expected result approximately 0.325751155, matching the official documentation’s example. Its standard zero-to-one midpoint is 0.5, but the distribution median is lower because its two shapes differ.
The mean of a standard beta distribution is Alpha/(Alpha+Beta), which is 1/3 for shapes five and ten. That mean is close to, but different from, the median. Do not replace an inverse calculation with the mean when a specification requests a percentile.
Errors and troubleshooting
Use positive shapes, increasing bounds and 0<p<1 for these worked quantiles. The dedicated help does not establish every endpoint behavior, error code, text coercion or extreme-tail tolerance; no such native behavior is certified here.
Keep probability cells and output-quantity cells separately labeled. If all outputs fall on an unexpected scale, inspect the bounds before changing the shape parameters. The function evaluates the supplied model; it does not estimate shapes, validate fit or assess uncertainty in the parameters.
Availability and verification
The exact function is documented in Collabora Office 26.04 desktop help. Its source snapshot was retrieved on 2026-10-13 and hash-verified for this guide on October 15, 2026. Desktop documentation is not proof of Online or mobile support.
First introduced release and last functioning version are unknown. The available dedicated source does not establish those releases. No native Collabora or WPS run occurred. Numerical values are independent mathematical expectations, not certification of application-specific errors, numerical tolerances or all edge cases. Formulas use English names and semicolon separators.
The help identifies the namespace COM.MICROSOFT.BETA.INV and says this function is outside the cited OpenFormula 1.3 standard. This format disclosure does not prove identical behavior in another product. Keep exact dotted names when checking imported workbooks.