Browsing by Author "CHEN, H"

Now showing 1 - 2 of 2

BOUNDARY-STRUCTURE DETERMINATION OF AG/SI(111) INTERFACES BY X-RAY-DIFFRACTION
(1995-07-15)
ABURANO, RD
;
HONG, H
;
ROESLER, JM
;
CHUNG, K
;
LIN, DS
Different Ag/Si(111) systems have been examined using synchrotron x-ray diffraction. Multi-atomic-layer deposition of Ag onto a Si(111)-(7X7) surface maintained at room temperature results in an unstrained, (111)-oriented film. The interface shows a Ag-modified (7X7) structure which when annealed above 200-250 degrees C transforms to a(1X1) structure. Although this is near the characteristic temperature for formation of the (root (3) over bar X root (3) over bar)R30 degrees surface reconstruction commonly observed for a monolayer of Ag adsorbed on Si(111), no evidence of this (root (3) over bar X root (3) over bar)R30 degrees reconstruction was found at the interface, A Ag monolayer (root (3) over bar X root (3) over bar)R30 degrees surface, further covered by multilayer Ag deposition at room temperature, also shows no indication of the (root (3) over bar X root (3) over bar)R30 degrees reconstruction at the interface. This indicates that the actual interface structure may or may not be related to the clean or adsorbed layer structures. The structure of the Ag-Si interface was further characterized by scans of the crystal truncation rods. Both the (7X7) interface prepared by room-temperature deposition and the annealed (1X1) interface show fairly sharp boundaries. The results suggest some intermixing occurs at the monolayer level for the annealed interface. The structure of the Ag film was also investigated.
DATA-DRIVEN EFFICIENT ESTIMATORS FOR A PARTIALLY LINEAR-MODEL
(1994-03-01)
CHEN, H
;
SHIAU, JJH
;
統計學研究所
;
Institute of Statistics
Chen and Shiau showed that a two-stage spline smoothing method and the partial regression method lead to efficient estimators for the parametric component of a partially linear model when the smoothing parameter is a deterministic sequence tending to zero at an appropriate rate. This paper is concerned with the large-sample behavior of these estimators when the smoothing parameter is chosen by the generalized cross validation (GCV) method or Mallows' C(L). Under mild conditions, the estimated parametric component is asymptotically normal with the usual parametric rate of convergence for both spline estimation methods. As a by-product, it is shown that the ''optimal rate' for the smoothing parameter, with respect to expected average squared error, is the same for the two estimation methods as it is for ordinary smoothing splines.