"FRACTIONAL REGRESSION MODELS" WEBSITE

 

Aim

In many regression models, the variable of interest is a proportion or a fraction, i.e. it is defined and observed only in the interval [0,1]. In Economics, examples include pension plan participation rates, firm market share, fraction of total weekly hours spent working, proportion of debt in the financing mix of firms, fraction of land area allocated to agriculture, and proportion of exports in total sales. The main aim of this page is to provide some references for these so-called fractional regression models and some simple code for computing the main alternative estimators and specification tests.

 

Main references

  • Seminal paper:

    Papke, L. and J.M. Wooldridge (1996), "Econometric methods for fractional response variables with an application to 401(K) plan participation rates", Journal of Applied Econometrics, 11(6), 619-32.

  • Survey on models, estimation methods and specification tests:

    Ramalho, 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.

  • Endogeneity and neglected heterogeneity:

    Ramalho, E.A. and J.J.S. Ramalho, "Moment-based estimation of nonlinear regression models with boundary outcomes and endogeneity, with applications to non-negative and fractional responses", Econometric Reviews, forthcoming.[Dataset: txt] [Code: R]

  • Panel data:

    Ramalho, E.A., J.J.S. Ramalho and L.M.S. Coelho, "Exponential regression of fractional-response fixed-effects models with an application to firm capital structure", Journal of Econometric Methods, forthcoming.[Data and code: zip]

  • DEA analyses:

    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. [Dataset: dta txt] [Code: Stata R]

  • Multivariate fractional data:

Murteira, J.M.R., J.J.S. Ramalho (2016), "Regression analysis of multivariate fractional data", Econometric Reviews, 35(4), 515-552.

Code

 

  • Functions / modules:

R Stata

Description

frm frm Estimation of one- and two-part fractional regression models. Available link functions:
(i) logit
(ii) probit
(iii) cauchit
(iv) loglog
(v) cloglog
frm.pe frm_pe

Average and conditional partial effects for:
(i) one-part models
(ii) the binary component of two-part models
(iii) the fractional component of two-part models
(iv) two-part models

frm.reset frm_reset RESET test for the link specification in: 
(i) one-part models
(ii) the binary component of two-part models
(iii) the fractional component of two-part models
frm.ggoff frm_ggoff GGOFF, GOFF1 and GOFF2 tests for the link specification in: 
(i) one-part models
(ii) the binary component of two-part models
(iii) the fractional component of two-part models
frm.ptest frm_ptest P test for assessing, against each other:
(i) two alternative specifications for the link function in one-part models
(ii) two alternative specifications for the link function in the binary component of two-part models
(iii) two alternative specifications for the link function in the fractional component of two-part models
(iv) two alternative specifications for two-part models
(v) a one-part model versus a two-part model
(vi) two models based on the same link specifiction(s) but using sets of non-nested regressors
frmhet --- Estimation of fractional regression models with neglected heterogeneity and/or endogenous covariates. Available estimators:
(i) GMMx

(ii) GMMxv
(iii) GMMz
(iv) LINx
(v) LINxv
(vi) LINz
(vii) QMLxv
frmhet.pe --- Average and conditional partial effects for models estimated by the command frmhet
frmhet.reset --- RESET test for models estimated by GMMx or LINx 
frmpd --- Estimation of panel data fractional regression models. Available estimators:
(i) GMMww
(ii) GMMc
(iii) GMMbgw
(iv) GMMpfe
(v) GMMcre
(vi) GMMpre
(vii) QMLcre

 

Questions

For any question about fractional regression models, please do not hesitate to contact me:

  • Joaquim J.S. Ramalho: