Regresión alzada y el número de condiciónalgunos problemas

  1. Salmerón Gómez, Román 1
  2. García García, Catalina 1
  3. García Pérez, José 2
  4. López Martín, María del Mar 1
  1. 1 Universidad de Granada
    info

    Universidad de Granada

    Granada, España

    ROR https://ror.org/04njjy449

  2. 2 Universidad de Almería
    info

    Universidad de Almería

    Almería, España

    ROR https://ror.org/003d3xx08

Journal:
Anales de ASEPUMA

ISSN: 2171-892X

Year of publication: 2016

Issue: 24

Type: Article

DIALNET GOOGLE SCHOLAR lock_openDialnet editor

More publications in: Anales de ASEPUMA

Abstract

Multiple linear regression analysis is a methodology widely applied in many different fields to establish relations between variables. When the independent variables present a high linear relation between them, the analysis is unstable and the conclusions may be questionable. This problem is known as approximate multicollinearity. The literature offers several options to solve this question being traditionally the most applied to eliminate the variables that are considered to cause multicollinearity. The raised regression is a quantitative technique that mitigates the problem of multicollinearity from a geometrical point of view. After applying this technique, it is recommendable to check if the collinearity has been mitigated or not. In this paper, it is presented the use of the condition number in the raise regression focusing on two problems that arise when this extension is addressed: the difficulty of obtaining a closed algebraic expression and the sensitivity of this measure to transformations in the original data.

Bibliographic References

  • Marquardt, D.W. (1970). “Generalized inverses, ridge regression, biased linear estimation and nonlinear estimation”. Technometrics, 12, 591–612. THEIL, H. (1971). “Principles of econometrics”. New York: Wiley.
  • Fox, J.; Monette, G. (1992). “Generalized collinearity diagnostics”. Journal of the American Statistical Association, 87, 178–183.
  • Belsley, D.A. (1982). “Assessing the presence of harmful collinearity and other forms of weak data through a test for signal-to-noise”. Journal of Econometrics, 20, 211–253.
  • Belsley, D.A., Kuh, E. y Welsch, R.E. (1980). “Regression diagnostics”. New York, NY: John Wiley.
  • Belsley, D.A. (1991). “Conditioning diagnostics: Collinearity and weak data in regression”. New York: John Wiley.
  • Hoerl, A.E.; Kennard, R.W. (1970a). “Ridge regression: Applications to nonorthogonal problems”. Technometrics, 12, 69–82. Hoerl, A.E.; Kennard, R.W. (1970b). “Ridge regression: Biased estimation for nonorthogonal problems”. Technometrics, 12, 55–67.
  • García, C. B., García, J.; Soto J. (2010). “The raise method: An alternative procedure to estimate the parameters in presence of collinearity”. Quality and Quantity, 45 (2), 403–423.
  • García, C. B., García, J., López Martín, M. M. y Salmerón, R. (2015a). “Collinearity: Revisiting the variance inflation factor in ridge regression”. Journal of Applied Statistics, 42 (3), 648–661.
  • García, C. B., García, J., Salmerón, R.; López Martín, M. M. (2015b). “Raise regression: selection of the raise parameter”. International Conference on Data Mining, 185–190.
  • García, J., Salmerón, R., García, C.B. y López Martín, M.M. (2015c). “Standardization of Variables and Collinearity Diagnostic in Ridge Regression”. International Statistical Review. DOI: 10.1111/insr.12099.
  • García, C., García, J., López Martín, M. M. y Salmerón, R. (2016). “Raise estimator: inference and properties”. Communications in Statistics-Theory and Methods. Pendiente de publicación.
  • Salmerón, R., García, J., López Martín, M.M. y García, C.B. (2016a). “Collinearity diagnostic applied in ridge estimation through the variance inflation factor”. Journal of Applied Statistics. DOI:10.1080/02664763. 2015.1120712.
  • Salmerón, R., García, J., García, C.B. y López Martín, M.M. (2016b). “Transformation of variables and the condition number in ridge estimation”. Computational Statistics and Data Analysis. En revisión.
  • Wissel, J. (2009). “A new biased estimator for multivariate regression models with highly collinear variables”. Ph.D. thesis, Dissertation zur Erlangung des naturwissenschaftlichen Doktorgrades der Bayerischen Julius-Maximilians-Universit¨atW¨urzburg.
Data source: Dialnet