FORECAST in Excel: Compatibility Status and Linear-Model Migration Checks

• Introduced: first introduction version is not established by the reviewed official history; the current product list is not a launch record.
• Availability: official help lists Excel for Microsoft 365, Excel for Microsoft 365 for Mac, Excel 2024, Excel 2024 for Mac, Excel 2021, Excel 2021 for Mac, Excel 2019, Excel 2016.
• Lifecycle: Microsoft recommends FORECAST.LINEAR, introduced as its replacement in Excel 2016. FORECAST remains available for backward compatibility; the page speaks of future deprecation and gives no retirement or last-supported date.
• Checked: October 14, 2026; official FORECAST reference. Documentation and independent models; no native Excel execution or wider platform certification.

=FORECAST(x,known_ys,known_xs)

All three arguments are required. They supply target X, known Y values and known X values, respectively. The official joint reference states that FORECAST and FORECAST.LINEAR have the same syntax and usage.

Microsoft’s wording anticipates eventual deprecation of FORECAST while explicitly saying it remains available. That is not evidence that the function has already stopped working or that a particular final supported version exists. Keep an existing workbook’s compatibility requirements visible when planning a migration.

Enter this table in A1:B5:

=FORECAST(50,B2:B5,A2:A5)

Expected result: 105. The observed values follow Y=2*X+5, so the fitted slope is 2 and intercept is 5. At X=50 the line evaluates to 2*50+5=105.

The means are X=25 and Y=55. Centered products total 1000, and centered X squares total 500. Their ratio confirms the slope. This exact-line fixture is useful for checking argument order before reviewing more complicated historical formulas.

=FORECAST.LINEAR(50,B2:B5,A2:A5)
=FORECAST(25,B2:B5,A2:A5)
=FORECAST.LINEAR(25,B2:B5,A2:A5)

Expected results: 105,55,55. The target 25 lies within the observed X range; 50 is beyond it. Both names evaluate the same documented line model for these ordinary numeric controls.

This comparison does not certify every historical Excel build, file conversion or floating-point boundary. Record the actual target versions and retain a copy of the original workbook when making a real migration.

Use X 1,2,3,4 and Y 2,1,4,3 in separate columns D2:E5:

=FORECAST(5,E2:E5,D2:D5)

Expected result: 4. The fitted slope is 0.6 and intercept is 1. The final observed Y is 3, and its preceding change is -1; simply extending that last change would give 2 instead. FORECAST uses all paired observations, not only the final two rows.

For this same table, target X=0 gives expected 1 and target X=2.5 gives 2.5. A fitted value at a chosen coordinate is not a guarantee about a future measurement.

In G2:G5 enter target X values 10,25,40,50 for the original exact-line table. Use:

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

Fill down. Expected results are 25,55,85,105. The fixed references preserve one model while the target varies. A migration review should compare both the function name and the referenced inputs; changing either range can change the result independently of the name.

1. Preserve the original formula and its observed input ranges.
2. Confirm the destination Excel version supports FORECAST.LINEAR.
3. Compare both names on representative complete numeric inputs, including an interior and an exterior target.
4. Inspect missing data, errors and unit conversions separately.

For the first fixture, multiplying all Y values by ten changes the forecast at 50 to 1050. Adding 100 to Y instead changes it to 205. Those are data transformations, not effects of switching function identities.

The joint reference assigns #VALUE! to nonnumeric target X, #N/A to empty or unequal observed series, and #DIV/0! to zero variance in known X.

Mixed-type observation handling and numerical extremes are not certified here. FORECAST.LINEAR is the recommended current spelling, while SLOPE and INTERCEPT expose its line coefficients. None of these functions establishes seasonality, causation or an uncertainty interval by itself.

• Microsoft FORECAST reference.
• Editorial testing standards.

Expected outputs are independent mathematical calculations or explicitly documented rules, not captured native Excel results. Examples use fictional data, English function names and comma separators. Native floating-point boundaries remain untested.