Summary
Estimation of orientation is a key operation at each step in projection pursuit. Since projection pursuit is a nonparametric algorithm, and since even low-dimensional approximations to the target function must converge to their limits at rates considerably slower than n 2-1 (where n is sample size), then it might be thought that the same is true of orientation estimates. It is shown in the present paper that this is not the case, and that estimation of orientation is a parametric operation, in the sense that, under mild nonparametric assumptions, correctly-chosen kernel-type orientation estimates converge to their limits at rate n 2-1 . This property is not enjoyed by standard projection pursuit orientation estimates, which converge at a slower rate than n 2-1 . Most attention in the present paper is focussed on the case of projection pursuit density approximation, but it is pointed out that our arguments hold generally. An important practical conclusion is that data should be smoothed less when estimating orientation than when constructing the final projection pursuit approximation.
References
Falk, M.: Relative efficiency and deficiency of kernel type estimates of smooth distribution functions. Statist. Neerlandica 37, 73–83 (1983)
Falk, M.: Relative deficiency of kernel type estimators of quantiles. Ann. Stat. 12, 261–268 (1984)
Friedman, J.H.: Exploratory projection pursuit. J. Am. Stat. Assoc. 82, 249–266 (1987)
Friedman, J.H., Stuetzle, W.: Projection pursuit regression. J. Am. Stat. Assoc. 76, 817–823 (1981)
Friedman, J.H., Stuetzle, W., Schroeder, A.: Projection pursuit density estimation. J. Am. Stat. Assoc. 79, 599–608 (1984)
Hall, P.: Asymptotic theory of minimum integrated square error for multivariate density estimation. In: Multivariate analysis IV, Krishnaiah, P.R. (ed.), pp. 289–309. Amsterdam: North Holland 1985
Hall, P.: On projection pursuit regression. Ann. Stat. to appear
Hall, P., Heyde, C.C.: Martingale limit theory and its application. New York: Academic Press 1980
Hoeffding, W.: Probability inequalities for sums of bounded random variables. J. Am. Stat. Assoc. 58, 13–30 (1963)
Huber, P.J.: Projection pursuit. (With discussion). Ann. Stat. 13, 435–475 (1985)
Jones, M.C., Sibson, R.: What is projection pursuit? J. Royal Stat. Soc. 150, 1–37 (1987)
Parzen, E.: On estimation of a probability density function and mode. Ann. Math. Stat. 33, 1065–1076 (1962)
Pollard, D.: Convergence of stochastic processes. New York Berlin Heidelberg: Springer 1984
Prakasa Rao, B.L.S.: Nonparametric functional estimation. New York: Academic Press 1983
Rosenblatt, M.: Remarks on some nonparametric estimates of a density function. Ann. Math. Stat. 27, 832–837 (1956)
Rosenblatt, M.: Curve estimation. Ann. Math. Stat. 42, 1815–1842 (1971)
Stone, C.J.: Optimal rates of convergence for nonparametric estimators. Ann. Stat. 8 1348–1360 (1980)
Stone, C.J.: Optimal global rates of convergence for nonparametric regression. Ann. Stat. 10, 1040–1053 (1982)
Stone, C.J.: An asymptotically optimal window selection rule for kernel density estimates. Ann. Stat. 12, 1285–1297 (1984)
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Hall, P. Estimating the direction in which a data set is most interesting. Probab. Th. Rel. Fields 80, 51–77 (1988). https://doi.org/10.1007/BF00348752
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF00348752
Keywords
- Stochastic Process
- Probability Theory
- Slow Rate
- Statistical Theory
- Density Approximation