LINEST in Collabora Office: Read Regression Coefficients and Fit Statistics
Use Collabora LINEST arrays, reproduce a complete five-row regression report, understand reversed multiple-predictor coefficients, and check residual sums.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: With X=1,2,3,4 and Y=2,3,5,4, LINEST’s independent fitted line is y=1.5+0.8x. A two-cell result row contains slope 0.8 followed by intercept 1.5.
Syntax and documented defaults
=LINEST(data_Y [; data_X [; linearType [; stats]]])
Data_Y is a single row or column of Y coordinates. Data_X is the corresponding X range; if omitted, the vendor explicitly defaults it to 1,2,…,n. Multiple predictors can be supplied in multiple matching rows or columns.
LinearType defaults to TRUE, permitting an intercept. FALSE forces the intercept to zero. Stats omitted or FALSE returns only the top coefficient row; TRUE returns the statistics table. Optional arguments should not be skipped while later positions are supplied under the page’s guidance, so the worked formulas specify all positions.
The documented method is least-squares linear regression. Multiple-predictor coefficients appear in reverse predictor order, with the intercept at the end.
Example: get slope and intercept
Enter A2:A5 as 1,2,3,4 and B2:B5 as 2,3,5,4. Select D2:E2 and enter as an array:
=LINEST(B2:B5;A2:A5;1;0)
The independent output row is 0.8 | 1.5. Predicted values at X=1,2,3,4 are 2.3,3.1,3.9,4.7. Observed minus fitted residuals are −0.3,−0.1,1.1,−0.7.
Enter the result as an array
The Collabora page documents array entry: select the intended output range, enter the formula, and confirm with Ctrl+Shift+Enter on the corresponding platform, or use the Function Wizard with Array checked. On platforms using Command, follow the Command/Ctrl shortcut indicated by the help. Do not type the outer array-formula braces yourself.
The category guidance warns that ordinary entry can return only the upper-left value. Existing array results are edited by selecting the whole array, not by overwriting one member cell. This guide does not assume modern automatic spill behavior from another spreadsheet.
Example: request the full statistics table
Select D8:E12 and enter =LINEST(B2:B5;A2:A5;1;1) as an array. Independent expectations are:
| Output row | Left cell | Right cell |
|---|---|---|
| Coefficients | slope 0.8 | intercept 1.5 |
| Coefficient standard errors | 0.424264069 | 1.161895004 |
| Fit summary | R² 0.64 | Y standard error 0.948683298 |
| Model test | F 3.555555556 | degrees of freedom 2 |
| Sums of squares | regression 3.2 | residual 1.8 |
These values are independent least-squares calculations, not captured Collabora results. The example deliberately has nonzero residual variation so the standard-error and F calculations are meaningful.
Advanced audit: reconstruct the sums of squares
The mean Y is 3.5. Squared deviations of observed Y from that mean sum to 5. Squared residuals sum to 1.8, leaving regression variation 3.2. Thus R²=3.2/5=0.64.
With four observations and two fitted coefficients, residual degrees of freedom are 2. Residual mean square is 1.8/2=0.9, so the Y standard error is √0.9≈0.948683298. The F statistic is 3.2/0.9≈3.555555556 for this one-predictor example.
SUMSQ can audit a residual column, and SQRT supplies the square root. A fit statistic is not a causal explanation or proof that the model will predict well outside the observed range.
Advanced example: read multiple-predictor order correctly
For an independent control, use six rows with X1 values 0,1,0,1,2,0, X2 values 0,0,1,1,0,2, and Y values 1,3,4,6,5,7. These exactly follow y=1+2X1+3X2.
With Data_X arranged as X1 then X2, the expected coefficient row is 3 | 2 | 1: X2’s slope first, then X1’s slope, then the intercept. This reverse order is explicitly documented in Collabora’s multiple-predictor example.
Label output columns immediately. Reading that row left to right as X1, X2, intercept would swap the effects while leaving plausible-looking numbers.
Errors and limitations
Match X and Y dimensions and keep observations paired. Select enough output cells for the requested table. Ordinary entry or a too-small output area can hide results; it does not mean LINEST returned only one statistic.
The source does not comprehensively enumerate coercion, rank-deficiency, or exact-fit error behavior. Perfectly duplicated predictors cannot independently identify all slopes, and an exactly fitted line can make some diagnostics degenerate. This guide does not assign native error codes or guarantee how every such case is represented. The statistics table’s explanatory scope is the documented arrangement and the explicit independent controls above.
Availability and verification
The exact function is documented in Collabora Office 26.04 desktop help, retrieved and reviewed on October 17, 2026. This establishes a documentation scope, not the first release that introduced the function or the last version in which it works; those versions and any retirement date remain unestablished. Collabora Online and mobile support are not inferred from desktop help, nor is behavior in another spreadsheet product.
The examples are documentation-based expectations with independent calculations. They have not been executed in Collabora Office. No application screenshot or native error/coercion certification is implied. Formulas use English function names and semicolon argument separators.
References
- Collabora Office 26.04: LINEST (checked October 17, 2026).
- Editorial testing standards.