Abstract
It becomes increasingly common to incorporate the predictors’ grouping knowledge into dimension reduction techniques. In this article, we establish a complete framework named groupwise sufficient dimension reduction via conditional distance clustering, when the grouping information is unknown. We introduce a simple-type conditional dependence measurement and a corresponding conditional independence test. A clustering procedure based on the measurement and test is constructed to detect the suitable group structure. Finally we conduct sufficient dimension reduction under the obtained structure. Both simulations and a real data analysis demonstrate that the clustering strategy is effective, and the groupwise sufficient dimension reduction method is generally superior to the classical sufficient dimension reduction method.
Similar content being viewed by others
References
Adragni KP, Al-Najjar E, Martin S, Popuri SK, Raim AM (2016) Group-wise sufficient dimension reduction with principal fitted components. Comput Stat 31:923–941
Cook RD, Weisberg S (1991) Disscussion of “Sliced inverse regression for dimension reduction,” by K. C. Li. J Am Stat Assoc 86:328–332
Cook RD (1996) Graphics for regressions with a binary response. J Am Stat Assoc 91:983–992
Cook RD (2007) Fisher lecture: dimension reduction in regression (with discussion). Stat Sci 22:1–26
Cook RD, Forzani L (2009) Likelihood-based sufficient dimension reduction. J Am Stat Assoc 104:197–208
Enz R (1991) Prices and earnings around the globe: a comparison of purchasing power in 48 cities. Union Bank of Switzerland, Economic Research Dept., Zurich
Ferré L (1998) Determining the dimension in sliced inverse regression and related methods. J Am Stat Assoc 93:132–140
Guo Z, Li L, Lu W, Li B (2015) Groupwise dimension reduction via envelope method. J Am Stat Assoc 110:1515–1527
Hastie TJ, Tibshirani RJ (1990) Generalized additive models. Chapman & Hall, London
Jaccard P (1912) The distribution of the flora in the alpine zone. New Phytol 11:37–50
Li KC (1991) Sliced inverse regression for dimension reduction. J Am Stat Assoc 86:316–342
Li B, Zha H, Chiaromonte F (2005) Contour regression: a general approach to dimension reduction. Ann Stat 33:1580–1616
Li Q, Racine JS (2007) Nonparametric econometrics: theory and practice. Princeton University Press, Princeton
Li B, Wang SL (2007) On directional regression for dimension reduction. J Am Stat Assoc 102:997–1008
Li L (2009) Exploiting predictor domain information in sufficient dimension reduction. Comput Stat Data Anal 53:2665–2672
Li L, Li B, Zhu LX (2010) Groupwise dimension reduction. J Am Stat Assoc 105:1188–1201
Székely GJ, Rizzo ML, Bakirov NK (2007) Measuring and testing dependency by correlation of distances. Ann Stat 35:2769–2794
Wand M, Jones M (1994) Multivariate plug-in bandwidth selection. Comput Stat 9:97–116
Wang X, Pan W, Hu W, Tian Y, Zhang H (2015) Conditional distance correlation. J Am Stat Assoc 110:1726–1734
Weisberg s (2015) R package “dr”
Xia Y, Tong H, Li WK, Zhu LX (2002) An adaptive estimation of dimension reduction space. J R Stat Soc Ser B 64:363–410
Yin X, Li B, Cook RD (2008) Successive direction extraction for estimating the central subspace in a multiple-index regression. J Multivar Anal 99:1733–1757
Acknowledgements
Funding was provided by MOE Project of Key Research Institute of Humanities and Social Sciences at Universities (Grant No. 16JJD910002) and Outstanding Innovative Talents Cultivation Funded Programs 2018 of Renmin Univertity of China.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest statement
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
Here we present proofs of theorems in this paper. In the proof of Theorem 1, the following lemma from Székely et al. (2007) is essential.
Lemma 1
If \(0<\alpha <2\), then for all x in \({\mathbb {R}}^d\)
where
and \(\varGamma (\cdot )\) is the complete gamma function. The integrals at 0 and \(\infty \) are meant in the principal value sense: \(\lim _{\varepsilon \rightarrow 0}\int _{{\mathbb {R}}^d\backslash \{\varepsilon B+\varepsilon ^{-1}B^c\}}\), where B is the unit ball (centered at 0) in \({\mathbb {R}}^d\) and \(B^c\) is the complement of B.
In this article, we take \(\alpha =1\), which is the simplest case, and the constant in Lemma 1 is
Proof of Theorem 1
Lemma 1 implies that there exist constants \(c_p\) and \(c_q\) such that for all \({\mathbf {X}}\in {\mathbb {R}}^p\), \({\mathbf {Y}}\in {\mathbb {R}}^q\),
For simplicity, consider the case \(p=q=1\). By plug in the empirical conditional characteristic functions with \((c_pc_q|\mathbf {t}|_p^{p+1}|\mathbf {s}|_q^{1+q})^{-1}=\pi ^{-2}t^{-2}s^{-2}\), the integration terms involve \(|\phi _{X,Y|Z}^n(t,s)|^2\), \(|\phi _{X|Z}^n(t)\phi _{Y|Z}^n(s)|^2\) and \(\overline{\phi _{X,Y|Z}^n(t,s)}\phi _{X|Z}^n(t)\phi _{Y|Z}^n(s)\).
For the first we have
where \(V_1=-\frac{1}{\sum \nolimits _{k=1}^nI_{Z_k}\sum \nolimits _{l=1}^nI_{Z_l}}i\sin (X_k-X_l)t\sin (Y_k-Y_l)sI_{Z_k}I_{Z_l}\) and the integral of \(V_1\) equals to zero.
Similarly, we have
and
where the integrals of \(V_2\) and \(V_3\) equal to zero.
Using the technique
note that the terms similar to
are cancelled in the final integration. So we only need to calculate integrals of the following type
For random vectors \({\mathbf {X}}\in {\mathbb {R}}^p\) and \({\mathbf {Y}}\in {\mathbb {R}}^q\), the same steps are applied. Thus Theorem 1 holds. \(\square \)
Proof of Theorem 2
Wang et al. (2015) have demonstrated that \({\mathcal {D}}_n({\mathbf {X}},{\mathbf {Y}}|{\mathbf {Z}}){\mathop {\rightarrow }\limits ^{a.s.}}{\mathcal {D}}({\mathbf {X}},{\mathbf {Y}}|{\mathbf {Z}})\) in the proof of their Theorem 4. Following Theorem 1, Theorem 2 naturally holds. \(\square \)
Proof of Theorem 3
(1) Let \(U_k=\exp (i\langle \mathbf {t},{\mathbf {X}}_k\rangle )-\phi ^n_{{\mathbf {X}}|{\mathbf {Z}}}(\mathbf {t})\) and \(V_k=\exp (i\langle \mathbf {s},{\mathbf {Y}}_k\rangle )-\phi ^n_{{\mathbf {Y}}|{\mathbf {Z}}}(\mathbf {s})\). Then by the Cauchy–Schwarz inequality,
Thus
This implies that \(0\le {\mathcal {R}}_n({\mathbf {X}},{\mathbf {Y}}|Z)\le 1\).
(2) Wang et al. (2015) have demonstrated that, for \(\mathbf {a}_1\in {\mathbb {R}}^p\) and \(\mathbf {a}_2\in {\mathbb {R}}^q\), \(b_1\ne 0\) and \(b_2\ne 0\), \(\mathbf {C}_1\in {\mathbb {R}}^{p\times p}\) and \(\mathbf {C}_2\in {\mathbb {R}}^{q\times q}\), \({\mathcal {D}}_n^2(\mathbf {a}_1+b_1\mathbf {C}_1{\mathbf {X}},\mathbf {a}_2+b_2\mathbf {C}_2{\mathbf {Y}}|Z)=|b_1b_2|{\mathcal {D}}_n^2({\mathbf {X}},{\mathbf {Y}}|Z)\). Following the definition of \({\mathcal {R}}_n({\mathbf {X}},{\mathbf {Y}}|Z)\) and Theorem 1, Theorem 3(2) naturally holds. \(\square \)
Proof of Theorem 4
Let \({\bar{\rho }}_n=\frac{1}{n}\sum _{i=1}^n\rho ({\mathbf {X}},{\mathbf {Y}}|Z=Z_i)\). Since \(E_Z[\rho ({\mathbf {X}},{\mathbf {Y}}|Z=Z_i)]={\mathcal {T}}\), \(Var_Z[\rho ({\mathbf {X}},{\mathbf {Y}}|Z=Z_i)]<\infty \) for \(\rho ({\mathbf {X}},{\mathbf {Y}}|Z=Z)\in [-1,1]\), following from the law of large numbers, we have that
Next we condition on Corollary 1 to prove that
Let \(R_{ni}\) denote \(R_n({\mathbf {X}},{\mathbf {Y}}|Z=Z_i)\), and \(\rho _i\) denote \(\rho ({\mathbf {X}},{\mathbf {Y}}|Z=Z_i)\). Conditioning on Corollary 1, and the property that almost sure convergence implies convergence in probability, we have that
and obviously
It follows that
and thus \(P(|{\mathcal {T}}_n-{\bar{\rho }}_n|\ge \varepsilon )\xrightarrow {n\rightarrow \infty }0\).
Since
Theorem 4
naturally holds. \(\square \)
Rights and permissions
About this article
Cite this article
Xu, X., Zhang, J. Groupwise sufficient dimension reduction via conditional distance clustering. Metrika 83, 217–242 (2020). https://doi.org/10.1007/s00184-019-00732-7
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00184-019-00732-7