INTERCEPT in Collabora Office: Fitted Values at Zero and Changed Coordinate Origins

• 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.

=INTERCEPT(data_y;data_x)

Both datasets are required. The Collabora section labels Y as dependent observations and X as independent observations. Its description does not establish a causal relationship between the variables.

The independent ordinary least-squares model uses intercept=mean Y-slope×mean X. These examples contain complete numeric pairs. Missing-data filtering, mixed-type conversions, native error codes and numerical-stability limits are not supplied by the compact source and remain unverified.

Enter X 2,4,6,8 in A2:A5 and Y 4,5,9,10 in B2:B5:

=INTERCEPT(B2:B5;A2:A5)

Expected result: 1.5. Mean X is 5 and mean Y is 7. Centered products 9,2,2,9 total 22; centered X squares total 20. The slope is 1.1.

The intercept is therefore 7-1.1×5=1.5. This predicts the line’s value at X=0, while the observed X values begin at 2. Whether that extrapolated point has a useful physical meaning depends on the variables being modeled.

In C2 enter and fill through C5:

=INTERCEPT($B$2:$B$5;$A$2:$A$5)+SLOPE($B$2:$B$5;$A$2:$A$5)*A2

Expected fitted values: 3.7,5.9,8.1,10.3. In D2 enter =B2-C2 and fill down. Expected residuals: 0.3,-0.9,0.9,-0.3.

The residuals sum to 0, but their squares total 1.8. Positive and negative errors cancel in a signed sum; zero residual sum is not proof that all observations were fitted exactly.

Create new X values in E2:E5 with =A2-5, filled down. They are -3,-1,1,3, with mean zero.

=INTERCEPT(B2:B5;E2:E5)

Expected intercept: 7. The slope remains 1.1, and the fitted Y values remain 3.7,5.9,8.1,10.3.

In the new coordinates, X-prime=0 means original X=5. The intercept changed because zero now represents the center of the observed X values, not the original origin.

This is why intercepts from models using different time origins or measurement baselines should not be compared without explaining those coordinate choices.

Add 10 to every original Y, producing 14,15,19,20. The expected intercept becomes 11.5, while the slope remains 1.1.

Double original Y instead, producing 8,10,18,20. Both coefficients double: intercept 3 and slope 2.2.

A common additive change in Y shifts the line vertically. A multiplicative unit change scales its vertical axis. Keep the transformation visible when comparing model coefficients from different reports.

For X 1,2,3 and Y 7,9,11, the exact line is Y=5+2×X, giving intercept 5 and slope 2.

Reversing both columns of the primary four-record table together retains intercept 1.5. Reordering only one variable can change the pairings and therefore the fitted line.

FORECAST.LINEAR evaluates a fitted line at an explicit target X. Its independent output at X=0 matches the intercept for complete numeric fits, but this is not a claim about every invalid-input edge case.

A constant X series gives no positive centered-X square total for a unique ordinary slope/intercept fit. The exact Collabora native error display is not documented in the section reviewed and remains unverified.

SLOPE reveals the rate-of-change coefficient, and RSQ describes squared correlation. Neither they nor INTERCEPT establish future outcomes, causation or an uncertainty interval. Keep the fitted range and model assumptions alongside the result.

• Collabora Office 26.04: INTERCEPT.
• 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.