Description
logOffset = ln(exposure); expectedCount = exposure*exp(linearPredictor).
Regression model terms depend on the response link, variable scale and the meaning of coefficients. A local marginal effect, elasticity or model mean computed from supplied parameters does not estimate those parameters or establish causality.
logOffset = ln(exposure); expectedCount = exposure*exp(linearPredictor).
When to use this tool
Use Count Model Exposure Log Offset Term when checking one stated regression-model contribution.
How to use the tool and how it works
- Enter linearPredictor, exposure. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read logOffset, expectedCount. Use the output convention below.
Require positive exposure. logOffset = ln(exposure); expectedCount = exposure*exp(linearPredictor).1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| linearPredictor | Log-link predictor excluding the separately supplied log exposure offset. | Required. Finite number. |
| exposure | Exposure | Required. Finite number; strictly above 0. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| logOffset | Log Offset | Required. Finite number. |
| expectedCount | Expected Count | Required. Finite number. |
{
"input": {
"linearPredictor": 0.6931471805599453,
"exposure": 3
},
"output": {
"logOffset": 1.0986122886681096,
"expectedCount": 6
}
}Limitations and assumptions
- The predictor excludes the exposure offset and exposure uses the modeled rate's denominator unit. This is a log-link mean term, not a fitted Poisson model, event probability or dispersion assessment.
- 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
Use a fitted model, suitable data and defensible identification for prediction, inference and causal interpretation.
References
Similar or alternative tools
- Elasticity Log Level One Point Term
Accept model log-level or level-level.
- Logit Marginal Effect One Covariate
effect = coefficient*probability*(1-probability).