Journal Information
Vol. 11. Issue 1.
Pages 24-32 (January - February 1997)
Vol. 11. Issue 1.
Pages 24-32 (January - February 1997)
Open Access
Actualizaciones en regresión: suavizando las relaciones
An update in regression: smoothing relationships
Visits
3811
E. Sánchez-Cantalejo Ramírez*, R. Ocaña-Riola
Escuela Andaluza de Salud Pública. Granada
This item has received

Under a Creative Commons license
Article information
Resumen

Una metodología muy utilizada al analizar distintos tipos de problemas de salud se basa en los modelos de regresión: lineal, logística, etc.; estos modelos dependen de un conjunto de parámetros que hay que estimar a partir de los datos del estudio. Sin embargo, tienen el inconveniente de ser muy rígidos en el sentido de imponer, en ocasiones, relaciones demasiado estrictas entre la variable resultado y las predic-toras. Los métodos de regresión no paramétrica presentan la ventaja de no establecer a priorí ninguna restricción, permitiendo así que los datos nos indiquen la forma funcional apropiada. En este artículo se presentan algunos métodos modernos de regresión no paramétrica que además de su utilidad per se sirven de inestimable ayuda en el proceso diagnóstico de los métodos de regresión paramétrica. La disponibilidad actual del software necesario debe posibilitar su mayor utilización, lo que redundará en una mejor comprensión de los problemas de salud estudiados.

Palabras clave:
Alisamiento
Regresión no paramétrica
Modelos aditivos generalizados
Summary

A frequently used methodology for the analysis of different kinds of health problems is based on regression models: lineal, logistic, etc.; these models depend on a set of parameters that must be estimated from the data. However, they present the drawback of being very rigid since, occasionally, they impose overly strict relations between the variables. Non-parametric regression methods present the advantage of not establishing a priori restrictions, allowing the data to indicate us the appropriate functional form. In this paper several modern non-parametric regression methods are presented that in addition to their usefulness per se can prone to be of invaluable help in the diagnostic process for parametric regression methods. The current availability of the necessary software should contribute to their increased use which, in turn, will probably lead to an improved understanding of the health problems under study.

Key words:
Smoothing
Non-parametric regression
Generalized additive models
Full text is only aviable in PDF
Bibliografía
[1.]
D. Birkes, Y. Dodge.
Alternative methods of regression.
[2.]
T.J. Hastie, R.J. Tibshirani.
Generalized additive models (with discussion).
Statist Sci, 1 (1986), pp. 297-318
[3.]
J.H. Friedman, W. Stuetzle.
Smoothing of scatterplots. Techi-nal Report Orion 003, Dept, of Statitstics.
Stanford University, CA, (1982),
[4.]
M.R. Segal, S.T. Weiss, F.E. Speizer, I.B. Tager.
Smoothing methods for epidemiologic analysis.
Statist Med, 7 (1988), pp. 601-611
[5.]
W. Hardle.
Smoothing techniques with implementation in S.
[6.]
W.S. Cleveland.
Robust locally weighted regression and smoothing scatterplots.
J Am Statist Ass, 74 (1979), pp. 828-836
[7.]
S. Greenland.
Invited commentary: a critical look at some popular meta-analytic methods.
Am J Epidemiol, 140 (1994), pp. 290-296
[8.]
N.E. Bush, M. Haberman, G. Donaldson, K.M. Sullivan.
Quality of life of 125 adults surviving 6-18 years after bone marrow transplantation.
Soc Sci Med, 40 (1995), pp. 479-490
[9.]
B.W. Silverman.
Some aspects of the spline smoothing approach to non-parametric regression curve fitting (with discussion).
J R Statist Soc B, 47 (1985), pp. 1-53
[10.]
J.r. Harrell FE, K.L. Lee, B.G. Pollock.
Regression models in clinical studies: determining relationship between predictors and response.
J Natl Cancer Inst, 80 (1988), pp. 1198-1202
[11.]
S. Durrleman, R. Simon.
Flexible regression models with cubic splines.
Statist Med, 8 (1989), pp. 551-561
[12.]
D.G. Royston R Altman.
Regression using fractional polynomials of continuous covariates: parsimonious parametric modelling (with discussion).
Appl Statist, 43 (1994), pp. 429-467
[13.]
P. Mc Cullagh, J.A. Nelder.
Generalized linear models.
2nd,
[14.]
T.J. Hastie, R.J. Tibshirani.
Generalized additive models.
[15.]
D.W. Hosmer, S. Lemeshow.
Applied logistic regression.
[16.]
A.A. Herman, T.J. Hastie.
An analysis of gestacional age neonatal size and neonatal death using non-parametric logistic regression.
J Clin Epidemiol, 43 (1990), pp. 1179-1190
[17.]
T.J. Hastie, R.J. Tibshirani.
Non-parametric logistic and pro-portional-odds regression.
Appl Statist, 36 (1987), pp. 260-276
[18.]
T.J. Hastie, R.J. Tibshirani.
Exploring the nature of covaria-tes effects in the proportional hazards model.
Biometrics, 46 (1990), pp. 1005-1006
[19.]
J. Schwartz.
Air pollution and hospital admissions for the elderly in Birmingham, Alabama.
Am J Epidemiol, 139 (1994), pp. 589-598
[20.]
D.G. Altman, S.N. Gooman.
Transfer of technology from statistical journals to the biomedical literature.
JAMA, 272 (1994), pp. 129-132
[21.]
B. Efron.
Computer-intensive methods in statistical regression.
SIAM Review, 30 (1988), pp. 421-449
[22.]
S-PLUS. Statistical Sciences Inc. Seattle WA, USA
[23.]
O.O. Aalen.
Further results on the non-parametric linear regression model in survival analysis.
Statistics in Medicine, 12 (1993), pp. 1569-1588
[24.]
A. Bowman, S. Young.
Graphical comparison of nonparame-tric curves.
Appl Statist, 45 (1996), pp. 83-98
[25.]
N. Galai, A. Muñoz, K. Chen, V.J. Carey, J. Chmiel, S.Y. Zhou.
Tracking of markers and onset of disease among HIV-1 sero-converters.
Statistics in Medicine, 12 (1993), pp. 2133-2145
Copyright © 1997. Sociedad Española de Salud Pública y Administración Sanitaria
Idiomas
Gaceta Sanitaria
Article options
Tools
es en

¿Es usted profesional sanitario apto para prescribir o dispensar medicamentos?

Are you a health professional able to prescribe or dispense drugs?