LOGEST in WPS Spreadsheet: Fit Exponential Coefficients
Estimate an exponential multiplier and constant, with positive-data examples and a corrected const interpretation.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer
For x=1,2,3 and y=6,12,24, =LOGEST(B2:B4,A2:A4,TRUE,FALSE) has an independent expected coefficient row of 2, 3. The result describes y=3×2^x: multiplier m=2 followed by constant b=3. LOGEST returns coefficients; GROWTH returns predicted y values.
Syntax and arguments
LOGEST(known_y's, [known_x's], [const], [stats])
Square brackets mark optional arguments and are not typed into a formula. known_y's is the required numeric response dataset. Use strictly positive y values for the exponential model in this guide. known_x's is optional and defaults to 1,2,3,… matching the response size. const is optional: TRUE or omission estimates b normally; FALSE fixes b=1. stats is optional: FALSE or omission returns coefficients and b, while TRUE requests additional regression statistics.
For one predictor, x and y have matching dimensions. For multiple predictors, y must be a single row or column. If y occupies one column, each x column is an independent variable; if y occupies one row, each x row is an independent variable. The source gives the coefficient order {mn,mn−1,…,m1,b}; preserve that reverse order when mapping multiple multipliers back to their predictors.
Worked worksheet: recover a doubling model
Enter 1,2,3 in A2:A4 and 6,12,24 in B2:B4. Reserve D2:E2 and enter =LOGEST(B2:B4,A2:A4,TRUE,FALSE) with the output-array workflow supported by your installed version. Expect 2 and 3 in the model’s coefficient row.
An independent fit of ln(y) against x gives slope ln(2) and intercept ln(3); exponentiation yields m=2 and b=3. These mathematical calculations are not observed WPS results. At x=4, the resulting model predicts 3×2^4=48. A multiplier of 2 means a 100% increase per x-unit; it does not mean a 2% increase or an increment of 2.
Advanced control: fixing b at one
Use y=2,4,8 in C2:C4 with the same x values:
=LOGEST(C2:C4,A2:A4,FALSE,FALSE)
The independent expectation is the coefficient row 2, 1. This is an exact y=2^x model. If instead you constrain the original 6,12,24 data, the model must refit its multiplier; it cannot preserve both the original multiplier and the new b=1 constraint.
The formal WPS parameter definition correctly specifies b=1 for FALSE. Its later walkthrough copies linear-regression dialog wording and incorrectly mentions b=0. This guide follows the formal exponential contract: b=0 would make y zero for every x and cannot describe the positive-data example.
Additional statistics and model interpretation
With stats=TRUE, the source lists coefficient/intercept standard errors, r², y-estimate standard error, F, degrees of freedom and regression/residual sums of squares. This guide intentionally does not promise a current-build full-array extraction layout or assign unverified numerical diagnostics to those cells. Check the installed help and array output before consuming them in downstream formulas. Exponential regression is understood through the logarithmic response model; do not casually interpret every diagnostic as original-unit linear-regression uncertainty.
Troubleshooting
Zero or negative y values are outside the positive-data exponential setup here. Select a suitable model rather than changing signs or adding arbitrary constants. Check paired observation rows, predictor variation and redundant columns. A single displayed coefficient may indicate an incomplete array-output workflow; do not read it as a predicted y value.
Growth is exponential in x, so changing the time unit changes the multiplier. Rescale consistently across both training data and prediction inputs. Even a perfect historical pattern can fail outside its fitted range; inspect the context before treating 48 as a forecast commitment. The official source does not establish current spill behavior across all platforms.
Availability and verification
The official WPS LOGEST lesson was checked on October 19, 2026. Its edition banner identifies WPS Office for Windows Personal. This Chinese first-party documentation is not proof of universal English-global, Mac, Linux, web or mobile support. Introduction and retirement versions remain unknown. All numerical results in this guide are independently calculated expectations, not native WPS test results or borrowed Excel outputs. Current GUI access was unavailable; no current-application screenshots or native execution are claimed.
Related guides
LN, EXP. Browse the WPS function library.