Syntax
STEYX(known_y, known_x)
| Parameter | Description |
|---|---|
| known_y | Parameter of the STEYX function. |
| known_x | Parameter of the STEYX function. |
Examples
Prediction precision
Formula
=STEYX(B2:B50, C2:C50)
Returns 3.2 — on average, actual y-values deviate about 3.2 units from the regression line's predictions.
Forecast confidence band
Formula
=FORECAST(15, B2:B50, C2:C50) + 2*STEYX(B2:B50, C2:C50)
Adds 2 standard errors to the forecast to get an approximate 95% upper confidence bound.
Comparing models
Formula
=STEYX(Y, X1) < STEYX(Y, X2)
Returns TRUE if X1 produces more precise predictions of Y than X2 does.
Common Errors
#N/A
Arrays have different lengths.
#DIV/0!
Fewer than 3 data points, or all x-values are the same.
Tips
In same units as y
STEYX is in the same units as your y-values. If y is dollars, STEYX is dollars of prediction error.
Approximate 95% confidence
Forecast +/- 2*STEYX gives a rough 95% prediction interval. For exact intervals, you'd need t-distribution critical values and leverage, but this is a practical approximation.
Needs at least 3 points
With only 2 points the regression line passes through both exactly, so STEYX would be 0 — but the formula requires 3+ points because there are no degrees of freedom with 2.
Try STEYX in Viztab
Import your data and use STEYX with 370+ other formulas. No signup required.
Open Viztab