A STUDY ON TESTS OF HYPOTHESIS BASED ON RIDGE ESTIMATOR

Abstract

The literature of Ridge regression include many articles that deal with point estimation of the coefficients vector . However, few of them tackle the statistical inference problem about or some of its components. One of them is introduced by Halawa and Basuiouni[1] who present non-exact tests based on Ridge regression by using two different biasing parameters (k) which are proposed by Hoerl and Kennard [2] and Hoerl et al. [3]. Thus, we investigate others popular k values used the Ridge regression for testing significance of regression coefficients. We compare tests in terms of type I error rates and powers by using Monte Carlo simulation. In addition, a real data example is presented.

Keywords

• Biasing factors,Ridge regression,Hypothesis testing,Type I error,Power of test

References

1. A.M. Halawa and M.Y.E. Basuiouni, “Tests of
2. Regression Coefficients under Ridge Regression
3. Models”, Journal of Statistical Computation and
4. Simulation, 65, 2000, pp. 341-356.
5. A.E. Hoerl and R.W. Kennard, “Ridge regression:
6. biased estimation for non-orthogonalproblems”,
7. Technometrics,12, 1970, pp. 55–67.
8. A.E. Hoerl, R.W. Kennard and K.F. Baldwin, “Ridge

9. regression: some simulation”, Communications in
10. Statistics, 5, 1975,pp. 105–123.
11. C.M. Theobald, “Generalizations of Mean Square
12. Error Applied to Ridge Regression”, Journal of the
13. Royal Statistical Society. Series B
14. (Methodolojical), 36, 1974,pp.103-106.
15. J.F. Lawless and P. Wang, “A Simulation Study of
16. Ridge and Other Regression Estimators”,
17. Communications in Statistics: Theory and
18. Methods,7, 1976, pp. 139-164.
19. A.P. Dempster, M. Schatzoff and N. Wermuth, “A
20. Simulation Study of Alternatives to Ordinary Least
21. Squares”, Journal of the American Statistical
22. Association, 72, 1977,pp. 77-91.
23. D.G. Gibbons, “A Simulation Study of Some Ridge
24. Estimators”, Journal of the American Statistical
25. Association, 76, 1981,pp.131-139.
26. A. K. Md. E. Saleh and B.M.G. Kibria,
27. “Performances of Some New Preliminary Test
28. Ridge Regression Estimators and Their Properties”,
29. Communicationsin Statistics: Theory and Methods,
30. , 1993,pp. 2747-2764.
31. B.M.G. Kibria, “Performance of Some New Ridge
32. Regression Estimators”, Communications in
33. Statistics: Simulation and Computation, 32,
34. ,pp. 419-435.
35. G. Khalaf and G. Shukur, “Choosing Ridge
36. Parameters for Regression Problems”,
37. Communications in Statistics: Theory and Methods,
38. , 2005,pp.1177-1182.
39. J. Zhang and M. Ibrahim, “A Simulation Study on
40. SPSS Ridge Regression and Ordinary Least
41. Squares Regression Procedures for Multicolliearity
42. data”, Journal of Applied Statistics, 32, 2005,pp.
43. -588.
44. A.K.Md.E. Saleh, “Theory of Preliminary Test and
45. Stein-Type Estimation with Applications”, Wiley,
46. New York, 2006.
47. M. Alkhamisi, G. Khalaf and G. Shukur, “Some
48. Modifications for Choosing Ridge Parameters”,
49. Communications in Statistics-Theory and Methods,
50. , 2006,pp. 2005-2020.
51. M. Alkhamisi and G. Shukur, “Developing Ridge
52. Parameters for SUR Model”, Communications in
53. Statistics -Theory and Methods, 37, 2008,pp. 544-
54. G. Muniz and B.M.G. Kibria, “On Some Ridge
55. Regression Estimators: An Empirical
56. Comparisons”, Communications in Statistics-
57. Simulation and Computation, 38, 2009,pp. 621-
58. F. Gökpınar and M. Ebegil, “A Comparative Study
59. on Ridge Estimators in Regression Problems”,
60. Journal of SainsMalaysiana, 2014 (in submission).
61. E.P. Liski, “A Test of the Mean Square Error
62. Criterion for shrinkage Estimators”,
63. Communications in Statistics: Theory and Methods,
64. , 1982,pp. 543-562.
65. E.P. Liski, “Choosing a shrinkage Estimator-a test
66. of the Mean Square Error Criterion”, Proc. First
67. Tampere Sem. Linear Models, 1983,pp. 245-262.
68. R.L. Obenchain, “Classical F-Tests and Confidence
69. Regions for Ridge Regression”, Technometrics, 19,
70. , pp. 429-439.
71. D. Coutsourides and C.G. Troskie, “F and t Tests
72. for a General Class of Estimators”, South African
73. Statistical Journal, 13, 1979,pp. 113-119.
74. A. Ullah, R.A.L. Carter and V.K. Srivastova, “The
75. Sampling Distribution of Shrinkage Estimators and
76. their F Ratio in Regression Model”, Journal of
77. Econometrics, 25, 1984,pp. 109-122.
78. M. Karakuş, A. Bayrak, E. Çalıkoğlu and M.
79. Kıralan, “Comparison of oxidation stability of
80. virgin olive oils from different locations of
81. Turkey”, ActaAlimentaria, 43, 2014, pp. 133-141.
82. H.Woods, H.H. Steinour, and H.R. Starke, “Effect
83. of composition of Portland cement on heat evolved
84. during hardening, Ind. Eng. Chem.”, 24, 1932, pp.
85. –1214.
86. M. Ebegil, F. Gökpınar, “A Test Statistic to Choose
87. Between Liu-Type and Least Squares Estimator
88. Based on Mean Square Error Criteria”, Journal of
89. Applied Statistics, 39(10), 2012, pp. 2081-2096.