RRI Function (Gnumeric)

• Introduced: Added in Gnumeric 1.9.11 release notes.
• Documented support: Registered in Gnumeric 1.12.62 in the Financial Functions plugin. No retirement is established.
• Last functioning version: Unverified for this guide. Source registration and independent arithmetic do not certify a running application version.
• Scope: English function names and comma-separated formulas. The cached source comes from the October 13–14, 2026 audit; the full guide and reference calculations were checked on October 15, 2026.

RRI solves (1+rate)^p=fv/pv, returning (fv/pv)^(1/p)-1. It is an endpoint growth calculation with no intervening deposits or withdrawals. The output is a rate per supplied period, not automatically a yearly return. Format a result such as 0.1 as a percentage to display 10%.

Function-specific help and complete worksheet callback establish the operation described here.

=RRI(p,pv,fv)

• p: Number of compounding periods. The callback rejects negative p and separately rejects zero.
• pv: Starting value; zero is rejected.
• fv: Ending value. Use same-sign nonzero endpoints for the ordinary growth examples. All three arguments are required.

The argument registration and callback were inspected together. A familiar function name in another spreadsheet is not a guarantee of the same input guards.

Put starting value 1000 in B2, ending value 1331 in B3 and period count 3 in B4. In B6 enter =RRI(B4,B2,B3); the reference result is 0.1, or 10% per period. In B7 enter =B2*(1+B6)^B4; this reconstructs 1331. That reverse check catches swapping the start and end cells or confusing three observations with three intervals.

If the values are measured at the beginning of year 1 and end of year 3, there are three annual intervals. If they are three equally spaced observations including both endpoints, there are only two intervals. Enter the actual elapsed compounding count rather than the number of rows.

These are independently calculated expected results from the pinned callback, with repeating or irrational values rounded to twelve decimal places. Error labels follow explicit source guards. They are not screenshots or native execution results.

Growth need not be positive. =RRI(2,1000,810) has reference rate -10%, because two 0.9 factors turn 1000 into 810. =RRI(1,1000,1100) gives 10% and unchanged endpoints give zero. Doubling over twelve periods gives approximately 5.94630943593% per period, not 100%/12.

For a five-year comparison, an RRI computed from beginning and ending values alone does not account for money added during those years. Use a cash-flow method when intermediate amounts matter. The source help permits fractional periods but warns that the rate interpretation is approximate; =RRI(0.5,1000,1100) has the algebraic reference rate 21% for a full unit period.

Negative p returns #NUM!; p equal to zero or pv equal to zero returns #DIV/0! in the inspected callback. A zero ending value with positive starting value and positive p gives the algebraic result -100%. For opposite-sign endpoints the power can fall outside the real-number domain. The callback has no explicit sign guard for that combination, and this guide does not certify the native error code or a meaningful financial interpretation.

Use same-sign values for the ordinary endpoint model. Unlike PV/FV cash-flow examples that conventionally give receipts and payments opposite signs, RRI needs a positive ratio for a straightforward real growth rate. Avoid replacing missing endpoints with zero: it either errors or represents a complete loss. A displayed 0.1 is already 10%, so do not multiply the formula by 100 and then apply percentage formatting.

Use actual numeric input cells for this fixture. Native handling of blank cells, numeric text, application-localized formula entry and display formatting was not executed here. Keep missing values distinct from deliberately supplied zero; using ISNUMBER to inspect inputs can prevent a text-formatted amount from being mistaken for a verified numeric fixture. The tables do not claim a particular result for untested coercions.

This guide uses the official pinned Gnumeric 1.12.62 source and independent reference arithmetic. It does not claim native execution, GUI validation, current-release parity or cross-product equivalence. The cached source date and the editorial checking date above describe separate activities. For an operational workbook, compare a small copy of this fixture with the installed version before relying on an imported schedule.

RATE, FV, PV.