PlanMaker FORECAST: Linear Estimates, Units and Extrapolation
Build PlanMaker linear forecasts with fixed data ranges, inspect residuals and distinguish least-squares predictions from local interpolation.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
FORECAST evaluates a least-squares straight line at a chosen x value. It uses all supplied observations. It does not simply select the nearest row or interpolate between the two neighboring observations.
Availability and version history
- Introduced: Documented by the PlanMaker 2010 manual; the first introduction release is not established.
- Availability: Documented in PlanMaker 2026; this is not a final working-version endpoint.
- Scope: The reference belongs to the 2026/NX manual family. Individual FreeOffice and mobile-build availability remains unverified.
- Checked: October 14, 2026, against the official function reference. Examples use vendor results and independent calculations; no native PlanMaker test was performed.
Syntax and argument order
=FORECAST(x, y_values, x_values)
| Argument | Meaning |
|---|---|
x |
Input value where you want a fitted response |
y_values |
Known dependent responses |
x_values |
Their corresponding independent inputs |
All three are required. The PlanMaker reference explicitly specifies least squares and warns that the known y values precede the known x values. Use complete numeric observations in corresponding rows. Formulas below use English names and comma separators.
Example: a fitted response at several input values
Create ResponseForecast. Put Input x and Observed y in A1:B1 and enter:
| Row | A: x | B: y |
|---|---|---|
| 2 | 1 | 3 |
| 3 | 2 | 5 |
| 4 | 3 | 4 |
| 5 | 4 | 8 |
Put Target x in G1 and Fitted y in H1. Enter 0, 2.5, 4 and 5 in G2:G5. In H2 enter:
=FORECAST(G2,$B$2:$B$5,$A$2:$A$5)
Fill H2 through H5. The dollar signs keep the known-data ranges fixed while the target cell changes.
| Row | Target x | Expected fitted y |
|---|---|---|
| 2 | 0 | 1.5 |
| 3 | 2.5 | 5 |
| 4 | 4 | 7.1 |
| 5 | 5 | 8.5 |
The prediction at x = 4 is 7.1, even though the observation there was 8. A best-fit line need not pass through each data point. Values at 0 and 5 also extend beyond the measured x range of 1–4.
Rebuild the line and inspect its residuals
In K2 enter =SLOPE(B2:B5,A2:A5); expect 1.4. In K3 enter =INTERCEPT(B2:B5,A2:A5); expect 1.5. The line is y = 1.5 + 1.4*x.
Put Fitted y, Residual and Squared residual in C1:E1. Enter =$K$3+$K$2*A2 in C2, =B2-C2 in D2 and =D2^2 in E2. Fill C2:E2 through row 5.
| Row | Fitted y | Residual | Squared residual |
|---|---|---|---|
| 2 | 2.9 | 0.1 | 0.01 |
| 3 | 4.3 | 0.7 | 0.49 |
| 4 | 5.7 | -1.7 | 2.89 |
| 5 | 7.1 | 0.9 | 0.81 |
In K5 enter =SUM(E2:E5); expect 4.2. In K6 enter =K3+K2*G3; expect 5, matching the FORECAST result at x = 2.5.
Straight-line interpolation using only the two observations at x = 2 and x = 3 would instead give 4.5 at their midpoint. FORECAST fits all four points, so its answer is 5. Choose a method because it matches the intended model, not because its output looks closest to a particular row.
Convert input units consistently
Suppose the original x measurements were in hours. Put Minutes in M1, enter =A2*60 in M2 and fill through M5.
- O2:
=FORECAST(150,B2:B5,M2:M5)gives 5, the same prediction as 2.5 hours. - O3:
=FORECAST(2.5,B2:B5,M2:M5)gives approximately 1.558333333. Here 2.5 is interpreted in minutes, inconsistent with the intended 2.5-hour target. - O4:
=SLOPE(B2:B5,M2:M5)gives approximately 0.023333333 response units per minute.
Changing the units of known x values requires changing the target x too. Merely changing the displayed number format does not perform a physical unit conversion.
Reproduce the official resistor forecast
Create ResistorForecast with Temperature C and Resistance ohms in A1:B1. Enter temperatures 8, 20, 25, 28 in A2:A5 and corresponding resistances 261, 508, 608, 680 in B2:B5.
Enter =FORECAST(15,B2:B5,A2:A5) in D2. The independent fitted result is 377258/931, approximately 405.218045113 ohms. It rounds to the manual’s 405.21805.
In D3 enter =INTERCEPT(B2:B5,A2:A5)+15*SLOPE(B2:B5,A2:A5) to reconstruct the same result. In D4 enter =FORECAST(0,B2:B5,A2:A5); expect approximately 93.698174006, agreeing with INTERCEPT as the source documents.
Practical limits
A straight-line estimate is not a guaranteed future value. The function does not introduce seasonality, choose a causal model or return a prediction interval. Inspect the observed pattern and residuals, especially when extrapolating.
The short reference does not give a comprehensive native error/coercion specification. Constant x values leave the ordinary regression slope undefined; mismatched, mixed-type or partially missing inputs need separate checking. No undocumented exact error token or cross-platform behavior is inferred here.