Skip to main content

Research Repository

Advanced Search

Outputs (5)

Constructing Economic Summary Indexes via Principal Curves (2010)
Presentation / Conference Contribution
Zayed, M., & Einbeck, J. (2010, December). Constructing Economic Summary Indexes via Principal Curves. Presented at COMPSTAT 2010, Paris, France

Index number construction is an important and traditional subject in both the statistical and the economical sciences. A novel technique based on localized principal components to compose a single summary index from a collection of indexes is propose... Read More about Constructing Economic Summary Indexes via Principal Curves.

Localized regression on principal manifolds (2010)
Presentation / Conference Contribution
Einbeck, J., & Evers, L. (2010, July). Localized regression on principal manifolds. Presented at 25th International Workshop on Statistical Modelling., Glasgow

We consider nonparametric dimension reduction techniques for multivariate regression problems in which the variables constituting the predictor space are strongly nonlinearly related. Specifically, the predictor space is approximated via ``local'' pr... Read More about Localized regression on principal manifolds.

Data compression and regression through local principal curves and surfaces (2010)
Journal Article
Einbeck, J., Evers, L., & Powell, B. (2010). Data compression and regression through local principal curves and surfaces. International Journal of Neural Systems, 20(3), 177-192. https://doi.org/10.1142/s0129065710002346

We consider principal curves and surfaces in the context of multivariate regression modelling. For predictor spaces featuring complex dependency patterns between the involved variables, the intrinsic dimensionality of the data tends to be very small... Read More about Data compression and regression through local principal curves and surfaces.

Data compression and regression based on local principal curves (2010)
Presentation / Conference Contribution
Einbeck, J., Evers, L., & Hinchliff, K. (2010, December). Data compression and regression based on local principal curves. Presented at 32nd annual Conference of the German Classification Society, Hamburg

Frequently the predictor space of a multivariate regression problem of the type y = m(x_1, …, x_p ) + ε is intrinsically one-dimensional, or at least of far lower dimension than p. Usual modeling attempts such as the additive model y = m_1(x_1) + … +... Read More about Data compression and regression based on local principal curves.