GROWTH Function (Gnumeric)

=GROWTH(known_ys,[known_xs],[new_xs],[affine])

• known_ys is the required collection of observed y-values.
• known_xs supplies the corresponding x-values. Omit it to use 1, 2, …, n.
• new_xs supplies the x-values to predict. Omit it to predict at the training x-values.
• affine defaults to TRUE. TRUE fits a multiplier b in y = b × m^x; FALSE fixes b to 1, which is equivalent to forcing the intercept of the log-transformed regression to zero.

Gnumeric’s implementation fits a single predictor and builds a one-column result with one row per collected new x-value. Do not infer multi-predictor support from LOGEST or from another spreadsheet’s GROWTH. Blank, text and Boolean entries are ignored by the numeric collectors; keep clean aligned numeric ranges rather than relying on omission to repair dirty data.

Enter this synthetic table without headers in the referenced cells:

Put 5 in D1 and 6 in D2. The formula =GROWTH(B1:B4,A1:A4,D1:D2) returns a two-row, one-column array with expected values 96 and 192. To inspect them individually, use =INDEX(GROWTH(B1:B4,A1:A4,D1:D2),1,1) and the same formula with row index 2. The exact model is y = 3 × 2^x, so the check is 3 × 32 = 96 and 3 × 64 = 192. Small floating-point rounding differences are normal.

Because A1:A4 already equals 1 through 4, =GROWTH(B1:B4,,D1:D2) gives the same expectations. Omitting new_xs instead gives the four fitted training values 6, 12, 24 and 48. These omitted-argument examples depend on that exact consecutive x-series; they are not interchangeable for dates or irregular measurements.

For equally spaced readings, use elapsed periods such as 0, 1, 2, 3 or 1, 2, 3, 4 consistently in both training and prediction ranges. If readings are unevenly spaced, supply elapsed days or hours explicitly. A new x-value of 5 means five units on that axis, not automatically the fifth future observation. Record the unit beside the input range and inspect fitted values before extrapolating far outside it. Exponential predictions can become enormous even when training data looks modest.

Unequal paired data lengths are documented as #NUM!. Empty usable prediction data and unsuccessful regression also lead to #NUM! in the inspected implementation. Use strictly positive numeric observations for the logarithmic fit; zero or negative y-values do not represent this model. Repeated identical x-values cannot identify an ordinary intercept-and-slope fit. If a prediction seems wrong, first verify which range is y and which is x, then check affine: FALSE changes the model, it does not merely hide the intercept in the output.

If only one answer is visible, inspect array entry or use the INDEX checks above. Do not copy an identical scalar formula down expecting it to choose successive result rows automatically.

The fn-stat module in the retained official Gnumeric 1.12.62 source registers this function. NEWS records the start of GROWTH implementation in 0.39 and reports one-dimensional GROWTH/TREND working in 0.42. These milestones distinguish initial development from working-support reports; historical execution was not repeated here, and the last functioning release is not established. This is a desktop worksheet function, with no web-edition claim. The official source archive was retained October 13–14, 2026 and inspected again October 17. It was not newly downloaded during this review.

The worked results are independently calculated expectations. On October 17, 2026, Gnumeric ssconvert 1.12.57 under C.UTF-8 recalculated the primary fixture successfully. The two INDEX-extracted predictions were 95.99999999999999 and 191.99999999999997, matching 96 and 192 within floating-point rounding. The native fixture used C1:C4 for y rather than the B1:B4 used in this guide; the values were identical. Other optional-argument, statistical-diagnostic and edge-case statements are source-derived or mathematical rather than separately native-tested here. The selected primary example is also shown in the genuine Gnumeric 1.12.57 screenshot below. No Gnumeric 1.12.62 execution or broader GUI workflow is claimed. Formula examples use English names and comma separators; adapt separators to the workbook locale. Array results are not a promise of Excel-style automatic spilling. Use Gnumeric array entry for a suitably sized result area, or INDEX to inspect individual elements.

Release history specifically records a GROWTH fix in Gnumeric 1.12.23; that is a maintenance milestone, not its introduction date.

• Official Gnumeric function catalog
• Pinned official statistical function help and implementation
• Pinned registration
• Official release history
• LN and EXP explain the log/exponential transformation.