Determinant efficiencies in ill-conditioned models. (English) Zbl 1229.62079
Summary: The canonical correlations between subsets of OLS estimators are identified with design linkage parameters between their regressors. Known collinearity indices are extended to encompass angles between each regressor vector and the remaining vectors. One such angle quantifies the collinearity of regressors with the intercept, of concern in the corruption of all estimates due to ill-conditioning. Matrix identities factorize a determinant in terms of principal subdeterminants and the canonical vector alienation coefficients between subset estimators – by duality, the alienation coefficients between subsets of regressors. These identities figure in the study of \(D\) and \(D_s\) as determinant efficiencies for estimators and their subsets, specifically, \(D_s\)-efficiencies for the constant, linear, pure quadratic, and interactive coefficients in eight known small second-order designs. Studies on \(D\)- and \(D_s\)-efficiencies confirm that designs are seldom efficient for both. Determinant identities demonstrate the propensity for \(D_s\)-inefficient subsets to be masked through near collinearities in overall \(D\)-efficient designs.
MSC:
| 62H20 | Measures of association (correlation, canonical correlation, etc.) |
| 62K05 | Optimal statistical designs |
| 62J05 | Linear regression; mixed models |
References:
| [1] | J. C. Kiefer, “Optimum experimental designs,” Journal of the Royal Statistical Society. Series B, vol. 21, pp. 272-319, 1959. · Zbl 0108.15303 |
| [2] | J. C. Kiefer, “Optimum designs in regression problems. II,” Annals of Mathematical Statistics, vol. 32, pp. 298-325, 1961. · Zbl 0099.13502 · doi:10.1214/aoms/1177705160 |
| [3] | J. C. Kiefer and J. Wolfowitz, “Optimum designs in regression problems,” Annals of Mathematical Statistics, vol. 30, pp. 271-294, 1959. · Zbl 0090.11404 · doi:10.1214/aoms/1177706252 |
| [4] | A. Wald, “On the efficient design of statistical investigations,” Annals of Mathematical Statistics, vol. 14, pp. 134-140, 1943. · Zbl 0060.30109 · doi:10.1214/aoms/1177731454 |
| [5] | M. J. Box and N. R. Draper, “Factorial designs, the |X\(^{\prime}\)X| criterion, and some related matters,” Technometrics, vol. 13, pp. 731-742, 1971. · Zbl 0228.62048 · doi:10.2307/1266950 |
| [6] | J. M. Lucas, “Which response surface design is best,” Technometrics, vol. 18, pp. 411-417, 1976. · Zbl 0342.62051 · doi:10.2307/1268656 |
| [7] | R. H. Myers, A. I. Khuri, and W. H. Carter, Jr., “Response surface methodology: 1966-1988,” Technometrics, vol. 31, no. 2, pp. 137-157, 1989. · Zbl 0693.62062 · doi:10.2307/1268813 |
| [8] | R. Sibson, “DA-optimality and duality,” in Progress in Statistics (European Meeting Statisticians, Budapest, 1972), Vol. II, J. Gani, K. Sarkadi, and I. Vincze, Eds., North-Holland, Amsterdam, The Netherlands, 1974. · Zbl 0305.62049 |
| [9] | S. D. Silvey, Optimal Design, Chapman & Hall, London, UK, 1980. · Zbl 0468.62070 |
| [10] | R. C. St. John and N. R. Draper, “D-optimality for regression designs: a review,” Technometrics, vol. 17, pp. 15-23, 1975. · Zbl 0295.62081 · doi:10.2307/1267995 |
| [11] | H. Dette, “A generalization of D- and D1-optimal designs in polynomial regression,” The Annals of Statistics, vol. 18, no. 4, pp. 1784-1804, 1990. · Zbl 0714.62068 · doi:10.1214/aos/1176347878 |
| [12] | H. Dette and T. Franke, “Robust designs for polynomial regression by maximizing a minimum of D- and D1-efficiencies,” The Annals of Statistics, vol. 29, no. 4, pp. 1024-1049, 2001. · Zbl 1012.62080 · doi:10.1214/aos/1013699990 |
| [13] | S. Huda, “On some Ds-optimal designs in spherical regions,” Communications in Statistics, vol. 20, no. 9, pp. 2965-2985, 1991. · Zbl 0800.62459 · doi:10.1080/03610929108830681 |
| [14] | S. Huda and A. A. Al-Shiha, “On D-optimal designs for estimating slope,” Sankhyá. Series A, vol. 61, no. 3, pp. 488-495, 1999. · Zbl 0972.62055 |
| [15] | C. M.-S. Lee, “D-optimal designs for polynomial regression when lower degree parameters are more important,” Utilitas Mathematica, vol. 34, pp. 53-63, 1988. · Zbl 0658.62086 |
| [16] | Y. B. Lim and W. J. Studden, “Efficient Ds-optimal designs for multivariate polynomial regression on the q-cube,” The Annals of Statistics, vol. 16, no. 3, pp. 1225-1240, 1988. · Zbl 0664.62075 · doi:10.1214/aos/1176350957 |
| [17] | W. J. Studden, “Ds-optimal designs for polynomial regression using continued fractions,” The Annals of Statistics, vol. 8, no. 5, pp. 1132-1141, 1980. · Zbl 0447.62070 · doi:10.1214/aos/1176345150 |
| [18] | D. M. Wardrop and R. H. Myers, “Some response surface designs for finding optimal conditions,” Journal of Statistical Planning and Inference, vol. 25, no. 1, pp. 7-28, 1990. · Zbl 0692.62062 · doi:10.1016/0378-3758(90)90003-D |
| [19] | H. P. Wynn, “The sequential generation of D-optimum experimental designs,” Annals of Mathematical Statistics, vol. 41, pp. 1655-1664, 1970. · Zbl 0224.62038 · doi:10.1214/aoms/1177696809 |
| [20] | H. P. Wynn, “Results in the theory and construction of D-optimum experimental designs,” Journal of the Royal Statistical Society. Series B. Methodological, vol. 34, pp. 133-147, 1972. · Zbl 0248.62033 |
| [21] | H. Dette, “On a mixture of the D- and D1-optimality criterion in polynomial regression,” Journal of Statistical Planning and Inference, vol. 35, no. 2, pp. 233-249, 1993. · Zbl 0768.62056 · doi:10.1016/0378-3758(93)90047-A |
| [22] | S. Huda, “On Ds-efficiency of some D-optimal designs,” Calcutta Statistical Association Bulletin, vol. 34, no. 133-134, pp. 119-122, 1985. · Zbl 0595.62071 |
| [23] | S. Huda and I. H. Khan, “On efficiency of some D-optimal designs in subset estimation,” Journal of the Indian Society of Agricultural Statistics, vol. 47, no. 2, pp. 151-164, 1995. |
| [24] | P. F. Schwartz, C. Gennings, L. K. Teuschler, and M. W. Farris, “Optimizing the precision of toxicity threshold estimation using a two-stage experimental design,” Journal of Agricultural, Biological, and Environmental Statistics, vol. 6, pp. 409-428, 2001. |
| [25] | S. L. Meadows-Shropshire, C. Gennings, W. H. Carter, and J. E. Simmons, “Analysis of mixtures of drugs/chemicals along a fixed-ratio ray without single-chemical data to support an additivity model,” Journal of Agricultural, Biological, and Environmental Statistics, vol. 9, pp. 500-514, 2004. |
| [26] | D. A. Belsley, “Centering, the constant, first-differencing, and assessing conditioning,” in Model Reliability, D. A. Belsley and E. Kuh, Eds., pp. 117-153, MIT Press, Cambridge, Mass, USA, 1986. |
| [27] | E. R. Mansfield and B. P. Helms, “Detecting multicollinearity,” American Statistician, vol. 36, pp. 158-160, 1982. |
| [28] | S. D. Simon and J. P. Lesage, “The impact of collinearity involving the intercept term on the numerical accuracy of regression,” Computer Science in Economics and Management, vol. 1, no. 2, pp. 137-152, 1988. · doi:10.1007/BF00427160 |
| [29] | D. A. Belsley, “Comment: well-conditioned collinearity indices,” Statistical Science, vol. 2, no. 1, pp. 86-91, 1987. · Zbl 0955.62605 · doi:10.1214/ss/1177013439 |
| [30] | H. Hotelling, “Relations between two sets of variates,” Biometrika, vol. 28, pp. 321-377, 1936. · Zbl 0015.40705 · doi:10.1093/biomet/28.3-4.321 |
| [31] | Å. Björck and G. H. Golub, “Numerical methods for computing angles between linear subspaces,” Mathematics of Computation, vol. 27, pp. 579-594, 1973. · Zbl 0282.65031 · doi:10.2307/2005662 |
| [32] | M. L. Eaton, Multivariate Statistics: A Vector Space Approach, John Wiley & Sons, New York, NY, USA, 1983. · Zbl 1160.62326 |
| [33] | G. W. Stewart, “Collinearity and least squares regression,” Statistical Science, vol. 2, no. 1, pp. 68-100, 1987. · Zbl 0955.62605 · doi:10.1214/ss/1177013439 |
| [34] | R. H. Myers, Classical and Modern Regression with Applications, PWS-KENT, Boston, Mass, USA, 2nd edition, 1990. |
| [35] | R. M. O’Brien, “A caution regarding rules of thumb for variance inflation factors,” Quality and Quantity, vol. 41, no. 5, pp. 673-690, 2007. · doi:10.1007/s11135-006-9018-6 |
| [36] | S. S. Wilks, Mathematical Statistics, John Wiley & Sons, New York, NY, USA, 1962. · Zbl 0173.45805 |
| [37] | H. Cramér, Mathematical Methods of Statistics, Princeton Mathematical Series, vol. 9, Princeton University Press, Princeton, NJ, USA, 1946. · Zbl 0063.01014 |
| [38] | K. G. Roquemore, “Hybrid designs for quadratic response surfaces,” Technometrics, vol. 18, pp. 419-423, 1979. · Zbl 0342.62052 · doi:10.2307/1268657 |
| [39] | H. O. Hartley, “Smallest composite designs for quadratic response surfaces,” Biometrics. Journal of the Biometric Society, vol. 15, pp. 611-624, 1959. · Zbl 0221.62029 · doi:10.2307/2527658 |
| [40] | G. E. P. Box and D. W. Behnken, “Some new three level designs for the study of quantitative variables,” Technometrics, vol. 2, pp. 455-475, 1960. · Zbl 0097.34406 |
| [41] | G. E. P. Box and K. B. Wilson, “On the experimental attainment of optimum conditions,” Journal of the Royal Statistical Society. Series B, vol. 13, pp. 1-45, 1951. · Zbl 0043.34402 |
| [42] | W. Notz, “Minimal point second order designs,” Journal of Statistical Planning Inference, vol. 6, no. 1, pp. 47-58, 1982. · Zbl 0473.62059 · doi:10.1016/0378-3758(82)90055-6 |
| [43] | A. T. Hoke, “Economical second-order designs based on irregular fractions of the 3n factorial,” Technometrics, vol. 16, pp. 375-384, 1974. · Zbl 0285.62052 · doi:10.2307/1267667 |
| [44] | M. J. Box and N. R. Draper, “On minimum-point second-order designs,” Technometrics, vol. 16, pp. 613-616, 1974. · Zbl 0292.62054 · doi:10.2307/1267614 |
| [45] | D. R. Jensen, “Efficiencies of some small second-order designs,” Statistics & Probability Letters, vol. 21, no. 4, pp. 255-261, 1994. · Zbl 0825.62658 · doi:10.1016/0167-7152(94)90135-X |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
