Ramsey RESET Test for Functional Form
Ramsey Regression Equation Specification Error Test (RESET) · Also known as: RESET test, regression specification error test, Ramsey RESET fonksiyonel form testi
The Ramsey RESET test, proposed by James Ramsey in 1969, is a general test for functional-form misspecification in a linear regression — for omitted nonlinear relationships between the response and the regressors. It adds powers of the fitted values to the model and checks whether they significantly improve the fit; if they do, the original linear specification has left systematic structure unexplained.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Apply the Ramsey RESET test after fitting a linear regression to screen for functional-form problems — omitted squared or interaction terms, a wrong link, or a needed transformation of the dependent variable. It is a useful, low-cost omnibus check that does not require specifying the alternative form. It cannot tell you what the correct specification is, only that something is wrong, and it can be confounded with other problems such as heteroskedasticity or omitted variables, so it should be read alongside those diagnostics. When RESET rejects, respond by adding polynomial or interaction terms, transforming variables, or moving to a more flexible model such as a GAM.
Strengths & limitations
- A general, low-cost screen for functional-form misspecification that needs no specific alternative.
- Easy to compute by augmenting the regression with powers of the fitted values.
- Sensitive to a wide range of nonlinearity and omitted-variable problems.
- Standard, widely implemented regression diagnostic with a familiar F-statistic.
- Indicates that something is wrong but not what the correct specification is.
- Can be confounded by heteroskedasticity or omitted variables, not just functional form.
- Power depends on the number and order of fitted-value powers included.
- A non-rejection does not prove the model is correctly specified.
Frequently asked
What does a significant RESET test indicate?
It indicates functional-form misspecification: powers of the fitted values add explanatory power, so the linear model has left systematic (likely nonlinear) structure unexplained. It does not tell you the correct form — only that the current one is inadequate.
How many powers of the fitted values should I add?
Usually the square and cube (ŷ² and ŷ³). Adding more increases the chance of detecting complex nonlinearity but risks collinearity and unstable estimates, so low orders are standard.
What should I do if RESET rejects?
Investigate the functional form: add quadratic or interaction terms, transform the dependent or independent variables (e.g., logs), or adopt a more flexible model such as a generalized additive model. Also check for heteroskedasticity and omitted variables, which can trigger a rejection.
Sources
- Ramsey, J. B. (1969). Tests for specification errors in classical linear least-squares regression analysis. Journal of the Royal Statistical Society: Series B, 31(2), 350–371. DOI: 10.1111/j.2517-6161.1969.tb00796.x ↗
How to cite this page
ScholarGate. (2026, June 2). Ramsey Regression Equation Specification Error Test (RESET). ScholarGate. https://scholargate.app/en/econometrics/ramsey-reset-test
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Multiple Linear RegressionStatistics↔ compare
- OLS RegressionEconometrics↔ compare
- Polynomial RegressionStatistics↔ compare
- White TestEconometrics↔ compare