FORECAST.LINEAR in Collabora Office: Fitted Lines with Actual X Spacing and Coordinate Changes
Evaluate Collabora linear forecasts with explicit paired data, preserve irregular X spacing and audit target transformations against a complete line model.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: FORECAST.LINEAR returns a fitted Y value for a supplied target X. It uses known paired Y and X observations to define a straight-line regression, rather than treating worksheet row numbers as the independent variable automatically.
Availability and version history
- Introduced: original Collabora release and launch date are not established by the reviewed sources.
- Availability: currently documented in Collabora Office 26.04 desktop help.
- Platform scope: online and mobile deployments are not inferred from this desktop-help release.
- Checked: October 14, 2026. Function-specific documentation and independent calculations; no native Collabora execution.
Syntax and the malformed source example
=FORECAST.LINEAR(value;data_y;data_x)
The formal Collabora reference requires three arguments. Its displayed example mistakenly separates the final range endpoints as B1;B50; a correct single range is B1:B50. The examples below follow the formal target/Y/X order and three-argument signature.
The independent model evaluates a+b×X, with ordinary least-squares intercept a and slope b. The source states the linear-regression purpose; it does not provide a complete native missing-data or error matrix.
Example 1: preserve irregular X spacing
Enter A1:B5:
| X | Y |
|---|---|
| 0 | 7 |
| 2 | 13 |
| 5 | 22 |
| 9 | 34 |
=FORECAST.LINEAR(6;B2:B5;A2:A5)
Expected result: 25. These records lie exactly on Y=7+3×X, giving 7+3×6=25. The unequal X gaps are part of the observations.
Replacing X with row positions 1,2,3,4 would describe different inputs. That alternative model has slope 9 and intercept -3.5; evaluating its target 6 would give 50.5, not 25. A row index and the measured X coordinate are not interchangeable.
Example 2: evaluate several targets consistently
Put targets 0,1,6,9,12 in D2:D6. In E2 enter and fill down:
=FORECAST.LINEAR(D2;$B$2:$B$5;$A$2:$A$5)
Expected values: 7,10,25,34,43. Targets 0 and 9 match observed X endpoints; 1 and 6 lie inside the observed interval; 12 extrapolates beyond it.
A formula can evaluate all five coordinates, but their practical forecasting reliability does not follow merely from successful arithmetic.
Advanced example: noise prevents exact point matching
Use X 0,1,2,3 and Y 4,5,9,10 in G2:H5:
=FORECAST.LINEAR(4;H2:H5;G2:G5)
Expected result: 12.5. The means are X=1.5 and Y=7; centered products total 11 and X squares total 5. Slope is 2.2 and intercept is 3.7.
The fitted value at observed X=2 is 8.1, while that row’s actual Y is 9. The function fits one line to all records rather than reproducing every noisy observation.
At mean X=1.5, the fitted result is 7, the mean Y. This supplies a useful arithmetic check for the intercept-containing model.
Advanced example: shift the X origin and the target together
Return to the original exact-line table. Add 100 to every observed X and change target 6 to 106. The slope stays 3, while the intercept becomes -293.
=-293+3*106
Expected fitted value remains 25. If the observed X values shift but the target incorrectly remains 6, the model instead yields -275. Consistent coordinate conversion requires transforming the target and the training values together.
For date or elapsed-time data, explain the origin and units explicitly rather than relying only on formatted labels.
Advanced checks: vertical transformations
Double every original Y, and the expected forecast at target 6 doubles to 50. Add 20 to original Y instead, and the forecast becomes 45. These changes follow the fitted line’s scale and vertical location.
Reverse both original columns together, and the target-6 result remains 25. Altering only one column’s order can break record pairing and produce a different fit.
Function identity and unresolved limits
The help identifies COM.MICROSOFT.FORECAST.LINEAR outside the cited OpenFormula 1.3 standard. It does not establish a first Collabora release or universal file-interchange behavior.
FORECAST remains separately documented in Collabora. Do not transfer Excel’s lifecycle wording or certify complete alias behavior without Collabora evidence. SLOPE and INTERCEPT explain the separately inspectable line coefficients.
Empty or unequal inputs, constant X, mixed types, source errors and extreme precision remain unverified native boundaries. A linear fit supplies neither seasonal structure nor a prediction interval, and extrapolation always needs a separate substantive justification.
References and verification
Complete function-specific documentation reviewed. Worked outputs are independent statistical models or explicitly documented rules, not application screenshots or native test results. Unverified type, error and numerical boundaries retain their stated scope.