RANDLOGNORM Function (Gnumeric)

• Introduction: Gnumeric 1.1.6 NEWS explicitly lists this new function. This release-history evidence is separate from a native test of that old version.
• Source support: Registered in the Gnumeric 1.12.62 random plugin. No retirement is established; the last functioning native application version is unverified.
• Evidence dates: Official Gnumeric source retained October 13–14, 2026 was reviewed October 15. Mathematical controls were independently recalculated October 15. No native application, compiled Gnumeric helper or random generator was executed.

Both arguments are required. The helper evaluates EXP(zeta+sigma*Z) using its standard-normal generator. Sigma must be nonnegative; a negative value produces #NUM!. Zeta can be negative, zero or positive. For sigma zero, the finite-result case is the constant EXP(zeta), although the helper still calls its normal generator before multiplying by zero.

For positive sigma, the ideal output has median EXP(zeta), mean EXP(zeta+sigma²/2), and variance (EXP(sigma²)-1)*EXP(2*zeta+sigma²). These are distinct quantities. The source’s help describes sigma broadly as a standard deviation; the implemented transformation makes clear that it is the standard deviation on the log scale.

Create LognormalControls. Put =LN(100) in B2 for zeta and 0.25 in B3 for sigma. In D2 enter =RANDLOGNORM($B$2,$B$3). D2 is a changing variate; no fixed expected draw is supplied.

In F2 enter =EXP(B2) and in F3 enter =EXP(B2+B3^2/2). Their independent reference results are 100 and approximately 103.17434075, the ideal median and mean. A single D2 result should not be expected to equal either statistic.

Put -2, -1, 0, 1 and 2 in A7:A11. In B7 enter =EXP($B$2+$B$3*A7) and fill through B11. These supplied normal-driver values give:

The table checks the parameter interpretation without pretending that Z takes only those five values. To check a negative zeta separately, =RANDLOGNORM(-1,0) has the source-derived constant EXP(-1), approximately 0.367879441171. =RANDLOGNORM(0,-0.25) is an error control, #NUM!.

Suppose an illustrative positive-valued model needs arithmetic mean 100 and standard deviation 20. Enter those numbers in H2 and H3. In H4 use =SQRT(LN(1+(H3/H2)^2)); in H5 use =LN(H2)-H4^2/2. H4 is approximately 0.198042200435, and H5 is approximately 4.58555982941.

The corresponding simulation formula is =RANDLOGNORM(H5,H4), with zeta first. Independently substituting those parameters into the ideal mean and variance formulas recovers 100 and 400. The median is approximately 98.0580675691, below the mean. This algebra assumes a strictly positive desired mean and nonnegative standard deviation; it is not an automatic fitting or suitability test for observed data.

For the original parameters, increasing zeta by LN(2) doubles every value for the same supplied Z. The ideal median becomes 200, and mean approximately 206.3486815. Increasing sigma instead changes the spread and also changes the arithmetic mean; it does not simply multiply all values by a constant.

If comparing two scenarios using the same underlying draw, compute one result in D2 and scale that cell. Two separate RANDLOGNORM calls are two generator calls and do not promise matching Z values. Repeating a reference to one random cell also does not create independent observations.

=RANDLOGNORM(0,0) gives 1 from the source arithmetic, while =RANDLOGNORM(LN(100),0) approximates 100 subject to the LN/EXP round trip. Zero sigma does not suppress the helper’s generator call, so it should not be described as leaving the shared random sequence untouched.

The ideal positive distribution does not justify a blanket promise that every finite-precision result is strictly positive: EXP can underflow to zero or overflow to an error for extreme inputs. The worked controls stay away from those extremes.

The function is registered as volatile. A recalculation can replace its result; neither a copied screenshot nor an earlier value is a promised next draw. There is no seed argument in this function’s worksheet signature. Exact recalculation triggers, shared-generator sequencing and saved-file replay were not tested.

The controls below separate a changing random formula from deterministic arithmetic. The supplied driver values are teaching inputs, not values captured from Gnumeric’s generator. Population summaries describe the ideal continuous distribution and are not empirical certifications of the finite-resolution implementation.

Use explicit finite numeric inputs for these controls. Text parsing, blank-cell coercion, locale-specific separators and array broadcasting are outside the verified examples. The numeric result constructor turns non-finite arithmetic into #NUM!; very large parameters or extreme transformed values can therefore fail even when the mathematical distribution is defined.

The underlying unit generator has finite resolution. Endpoint exclusions, rounding, overflow and underflow mean a theoretical continuous distribution is not a literal list of every machine output. No random samples, native worksheet execution, distribution-quality test, GUI operation or cross-application file transfer was performed for this guide. The executable checks verify deterministic transformations and mathematical reference values only.

• Function help and worksheet callback
• Pinned transformation helper, finite-resolution unit generator, and volatile registrations
• Finite-number result and error conversion
• Normal helper and cached companion draw
• NIST lognormal parameters and moments
• Gnumeric official function catalogue

RANDWEIBULL, RANDLAPLACE.