Description
Convert a supplied log-linear coefficient change to a model-implied multiplicative and percentage change.
Econometric calculations separate supplied residuals, regressors, weights, restrictions, panel groups and reference periods. One score or moment contribution is not a fitted estimator, and one scalar covariance term is not a standard error. Unit rescaling and event-time arithmetic preserve conventions without establishing model identification or causal interpretation.
Convert a supplied log-linear coefficient change to a model-implied multiplicative and percentage change.
When to use this tool
Use Log Linear Coefficient Percent Crosswalk when you already have logCoefficient, predictorChange and need to convert a supplied log-linear coefficient change to a model-implied multiplicative and percentage change.
How to use the tool and how it works
- Enter logCoefficient, predictorChange. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read exactMultiplicativeFactor, exactPercentChange. Use the output convention below.
Let x=logCoefficient*predictorChange; return exp(x) and 100*expm1(x), using the natural-log convention.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| logCoefficient | Log Coefficient | Required. Finite number. |
| predictorChange | Predictor Change | Required. Finite number. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| exactMultiplicativeFactor | Exact Multiplicative Factor | Required. Finite number. |
| exactPercentChange | 100*expm1(logCoefficient*predictorChange); natural-log model change, not statistical certainty. | Required. Finite number. |
{
"input": {
"logCoefficient": 0.1,
"predictorChange": 2
},
"output": {
"exactMultiplicativeFactor": 1.2214027581601699,
"exactPercentChange": 22.140275816016985
}
}Limitations and assumptions
- Exact refers to exponentiation within the supplied model, not causal or statistical certainty. No retransformation bias or uncertainty correction is added; extreme negative x can underflow to factor zero and -100 percent.
- Inputs must satisfy both the field types and the cross-field conditions above. Invalid input causes an error; a successful false flag or documented placeholder is a different result.
- Numeric calculations use binary floating point unless exact decimal-integer arithmetic is explicitly stated. No hidden measurement tolerance or decimal rounding is added; nonfinite or out-of-contract results are rejected. Numeric signed-zero results are exported as zero.
Alternative or Complementary methods
Fit and evaluate the full specified model with appropriate data, grouping and uncertainty treatment for inference. Aggregate the required score/moment terms and use the complete covariance calculation; a local residual, cutoff flag or supplied coefficient transformation does not validate the estimator.
References
Similar or alternative tools
- Regression Prediction Retransformation Factor
Apply a supplied positive smearing factor to an exponentiated log prediction.
- Regression Coefficient Unit Rescale
Rescale a supplied regression coefficient between predictor and outcome units.