Abstract
In this paper we consider a regression model and a general family of shrinkage estimators of regression coefficients. The estimation of each individual regression coefficient is important in some practical situations. Thus, we derive the formula for the mean squared error (MSE) of the general class of shrinkage estimators for each individual regression coefficient. It is shown analytically that the general family of shrinkage estimators is dominated by its positive-part variant in terms of MSE whenever there exists the positive-part variant or, in other words, the shrinkage factor can be negative for some parameter and data values.
Similar content being viewed by others
References
Baranchik AJ (1970) A family of minimax estimators of the mean of a multivariate normal distribution. Ann Math Stat 41:642–645
Efron B, Morris C (1973a) Combining possibly related estimation problems (with discussion). J R Stat Soc B 35:379–421
Efron B, Morris C (1973b) Stein’s estimation rule and its competitors—an empirical Bayes approach. J Am Stat Assoc 68:117–130
Goldstein M, Khan MS (1985) Income and price effects in foreign trade. In: Kenen B, Jones RW (eds) Handbook of international economics, vol 2. Elsevier, Amsterdam, pp 1041–1105
James W, Stein C (1961) Estimation with quadratic loss. In: Proceedings of the fourth Berkeley symposium on mathematical statistics and probability, vol 1. University of California Press, Berkeley, pp 361–379
Judge GG, Bock ME (1978) The statistical implications of pre-test and Stein-rule estimators in econometrics. Elsevier, North-Holland
Judge GG, Yancey TA (1986) Improved methods of inference in econometrics. North-Holland, Amsterdam
Namba A (2003a) Dominance of the positive-part shrinkage estimator in a regression model when relevant regressors are omitted. Stat Probab Lett 63:375–385
Namba A (2003b) On the use of the Stein variance estimator in the double \(k\)-class estimator when each individual regression coefficient is estimated. Stat Pap 44:117–124
Namba A, Ohtani K (2002) MSE performance of the double \(k\)-class estimator of each individual regression coefficient under multivariate \(t\)-errors. In: Ullah A, Wan AT, Chaturvedi A (eds) Handbook of applied econometrics and statistical inference. Marcel Dekker, New York, pp 305–326
Namba A, Ohtani K (2010) Risk performance of a pre-test ridge regression estimator under the LINEX loss function when each individual regression coefficient is estimated. J Stat Comput Simul 80:255–262
Nickerson DM (1988) Dominance of the positive-part version of the James–Stein estimator. Stat Probab Lett 7:97–103
Ohtani K (1997) Minimum mean squared error estimation of each individual coefficient in a linear regression model. J Stat Plan Inference 62:301–316
Ohtani K, Kozumi H (1996) The exact general formulae for the moments and the MSE dominance of the Stein-rule and positive-part Stein-rule estimators. J Econom 74:273–287
Rao CR, Shinozaki N (1978) Precision of individual estimators in simultaneous estimation of parameters. Biometrika 65:23–30
Stein C (1956) Inadmissibility of the usual estimator for the mean of a multivariate normal distribution. In: Proceedings of the third Berkeley symposium on mathematical statistics and probability, vol 1, University of California Press, Berkeley, pp 197–206
Uemukai R (2011) Small sample properties of a ridge regression estimator when there exist omitted variables. Stat Pap 52:953–969
Ullah A (1974) On the sampling distribution of improved estimators for coefficients in linear regression. J Econom 2:143–150
Acknowledgments
The author is grateful to Kazuhiro Ohtani for his helpful comments and suggestions. He is also grateful to Aman Ullah, Tae-Hwy Lee and other participants of the econometric seminar held at University of California, Riverside on May 12, 2010. Also, he would like to thank the anonymous referees and the editors for their very useful comments and suggestions. This work was supported by JSPS KAKENHI Grant Numbers 23243038.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
In this appendix, we derive the formulae for \(H(p,q;c)\) and \(J(p,q;c)\). First, we derive the formula for \(H(p,q;c)\). Let \(u_1=(h'\widehat{\gamma })^2/\sigma ^2\), \(u_2=\widehat{\gamma }'(I_k -hh')\widehat{\gamma }/\sigma ^2\) and \(u_3=e'e/\sigma ^2\). Then, \(u_1 \sim \chi _1^{\prime 2}(\lambda _1)\) and \(u_2 \sim \chi _{k-1}^{\prime 2}(\lambda _2)\), where \(\chi _f^{\prime 2}(\lambda )\) is the noncentral chi-square distribution with \(f\) degrees of freedom and noncentrality parameter \(\lambda \), \(\lambda _1=(h'\gamma )^2/\sigma ^2\) and \(\lambda _2=\gamma '(I_k - hh')\gamma /\sigma ^2\). Further, \(u_3\) is distributed as the chi-square distribution with \(\nu =n-k\) degrees of freedom, and \(u_1\), \(u_2\) and \(u_3\) are mutually independent.
Using \(u_1\), \(u_2\) and \(u_3\), \(H(p,q;c)\) is expressed as
where
\(w_i(\lambda )=\exp (-\lambda /2)(\lambda /2)^i/i!\), and \(R\) is the region such that \((u_1+u_2)/u_3\ge c\).
Making use of the change of variables, \(v_1=(u_1+u_2)/u_3\), \(v_2=u_1 u_3/(u_1+u_2)\) and \(v_3=u_3\), (25) reduces to
Again, making use of the change of variable, \(z_1=v_2/v_3\), (27) reduces to
Further, making use of the change of variable, \(z_2=v_3(v_1+1)/2\), (28) reduces to
Finally, making use of the change of variable, \(t=v_1/(1+v_1)\). we obtain (14) in the text.
Next, we derive the formula for \(J(p,q;c)\). Differentiating \(H(p,q;c)\) given in (14) with respect to \(\gamma \), we have
where we define \(w_{-1}(\lambda _1) = w_{-1}(\lambda _2) =0\). Since \(h'h=1\), we obtain
Expressing \(H(p,q;c)\) by \(\widehat{\gamma }\) and \(e'e\), we have
where \(F=(\widehat{\gamma }'\widehat{\gamma })/(e'e)\), \(f(e'e)\) is the density function of \(e'e\), and
is the density function of \(\widehat{\gamma }\).
Differentiating \(H(p,q;c)\) given in (32) with respect to \(\gamma \), and multiplying \(h'\) from the left, we obtain
Rights and permissions
About this article
Cite this article
Namba, A. MSE dominance of the positive-part shrinkage estimator when each individual regression coefficient is estimated. Stat Papers 56, 379–390 (2015). https://doi.org/10.1007/s00362-014-0586-6
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-014-0586-6