help forfrm_reset-------------------------------------------------------------------------------

Title

frm_reset-- RESET test for fractional regression models

Syntax

frm_reset[anything] [,options]

optionsDescription -------------------------------------------------------------------------lastpower(#)specify the maximum power of the linear predictor to be used in the RESET testlmspecify the LM version of the RESET test to be performedwaldspecify the Wald version of the RESET test to be performedlrspecify the LR version of the RESET test to be performedmaximize_optionscontrol the maximization process of the estimations required to perform Wald and LR tests -------------------------------------------------------------------------where

anything, if provided, is the name under which estimation results were saved via estimates store. Otherwise,frm_resetis applied to the last estimation results, even if these were not already stored.

Description

frm_resetapplies the RESET test statistic to fractional regression models estimated via frm.frm_ptestmay be used to test the link specification of: (i) one-part fractional regression models; (ii) the binary component of two-part fractional regression models; and (iii) the fractional component of two-part fractional regression models. See Ramalho, Ramalho and Murteira (2011) for details on the application of the RESET test in the fractional regression framework.

Options

lastpower(#)} specify the maximum power of the linear predictor to be used in the RESET test. The default option islastpower(3), which implies that two RESET statistics are computed, one using quadratic powers of the fitted values as test variable and the other considering quadratic and cubic powers of the fitted values as test variables.

lmspecifies that the LM version of the RESET test is to be performed. Unless the model to be tested is the binary component of a two-part fractional regression model, a robust version is implemented. This is the default option.

waldspecifies that the Wald version of the RESET test is to be performed. It may be implemented a standard, a robust or a cluster version of the Wald test, depending on how the base model was estimated.

lrspecifies that the LR version of the RESET test is to be performed. This option is only available for the binary component of a two-part fractional regression model.

maximize_options:technique(algorithm_spec),iterate(#),trace,difficult,gradient,showstep,hessian,showtolerance,tolerance(#),ltolerance(#),nrtolerance(#),from(init_specs); see[R] maximizefor details.

ExamplesSetup - data used in Ramalho, Ramalho and Henriques (2010)

. use http://evunix.uevora.pt/~jsr/stata/JPA-2010.dtaTesting the logit specification of a standard fractional regression model

. frm SCORE LANDLORD LIVESTOCK CROP SIZE SUBSIDIES ALTO CENTRAL BAIXO. frm_resetTesting probit and cloglog specifications of binary regression models using LR-based RESET tests with quadratic, cubic and fourth powers of the linear predictor

. frm SCORE LANDLORD LIVESTOCK CROP SIZE SUBSIDIES ALTO CENTRALBAIXO, m(2Pbin) linkb(probit) inf(1). estimates store a1. frm SCORE LANDLORD LIVESTOCK CROP SIZE SUBSIDIES ALTO CENTRALBAIXO, m(2Pbin) linkb(cloglog) inf(1). estimates store a2. frm_reset a1, lr l(4). frm_reset a2, lr l(4)

Saved results

frm_resetsaves results of the following type inr():Scalars

r(LM2)LM-based RESET statistic that uses only quadratic powers of the linear predictor as test variabler(LM2p)p-value for the statisticr(LM2)If

lastpower(#)is higher than 2, thenr(LM3),r(LM3p), etc. are also saved. If Wald and LR versions of the RESET test are computed, thenr(W2),r(LR2), etc. are also saved.

AuthorJoaquim J.S. Ramalho Department of Economics University of Evora Portugal jsr@uevora.pt

Remarks

frm_resetis not an official Stata command. For further help and support, please contact the author. Please notice that this software is provided as is, without warranty of any kind, expressed or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and noninfringement. In no event shall the author be liable for any claim, damages, or other liability, whether in an action of contract, tort, or otherwise, arising from, out of, or in connection with the software or the use or other dealings in the software.

ReferencesRamalho, E.A., J.J.S. Ramalho and J.M.R. Murteira (2011), "Alternative estimating and testing empirical strategies for fractional regression models", Journal of Economic Surveys, 25(1), 19-68.

Ramalho, E.A., J.J.S. Ramalho and P.D. Henriques (2010), "Fractional regression models for second stage DEA efficiency analyses", Journal of Productivity Analysis, 34(3), 239-255.

Also see