FORECAST in Collabora Office: Linear Fitted Values and a Documented Syntax Correction

• 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.

=FORECAST(value;data_y;data_x)

All three arguments are required. The formal Collabora syntax is unambiguous, but its displayed example splits the intended final range into B1;B50, creating an inconsistent extra argument. Use a proper range such as B1:B50; this guide follows the documented three-argument signature.

The worked figures use the independent ordinary least-squares model, intercept+slope×target X, with complete numeric pairs. The compact help does not supply every missing-data, type or native error rule.

Enter A1:B5:

=FORECAST(9;B2:B5;A2:A5)

Expected result: 20. The observations lie exactly on Y=2+2×X. The means are X=4 and Y=10; centered products total 40, and centered X squares total 20, confirming slope 2 and intercept 2.

The target 9 lies beyond the observed X range 1 through 7, so this is an extrapolation. A mathematically continued line does not establish that a real process will continue the same way.

=FORECAST(4;B2:B5;A2:A5)

Expected result: 10. Target 4 lies within the observed interval and equals mean X. The fitted line passes through the pair of numeric means in this ordinary regression model.

An interior target is still a fitted estimate. A noisy dataset can produce a fitted value different from an actual observation at the same X.

Enter X 0,1,2,3 in D2:D5 and Y 1,3,2,6 in E2:E5:

=FORECAST(4;E2:E5;D2:D5)

Expected result: 6.5. The fitted line is Y=0.9+1.4×X. Extending only the final observed change from 2 to 6 would instead give 10, which is a different calculation.

For the noisy table, target 1.5 gives 3, the fitted mean Y; target 0 gives 0.9, not the observed first Y of 1. The line is fitted globally rather than required to pass through every point.

Return to the original exact-line table. Put targets 0,2,4,9 in G2:G5. In H2 enter and fill down:

=FORECAST(G2;$B$2:$B$5;$A$2:$A$5)

Expected results: 2,6,10,20. The absolute references hold the training data fixed while the target changes.

Targets 0 and 9 are outside the observed interval. Label extrapolations rather than presenting all rows as equally supported by the observed X range.

Double every original Y and evaluate target 9 again. Expected forecast becomes 40. Add 100 to every original Y instead, and the forecast becomes 120.

Reverse both original columns together, and the expected target-9 result remains 20. Sorting or filtering one column independently can alter the pairs and the fitted line. Keep an observation identifier alongside both variables.

FORECAST remains explicitly documented in Collabora Office 26.04. FORECAST.LINEAR has its own section with the same stated linear-regression purpose. Another product’s replacement or deprecation policy does not establish Collabora retirement or complete native equivalence.

SLOPE and INTERCEPT expose the independently modeled coefficients. None supplies seasonality, a causal conclusion or an uncertainty interval by itself.

Constant-X data leave the ordinary slope denominator zero. Exact native errors for that case, empty or unequal series, text/Boolean conversion and numerical extremes remain unverified here rather than imported from Excel.

• Collabora Office 26.04: FORECAST.
• Editorial testing standards.

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.