Fredholm integral relation between compound estimation and prediction (FIRCEP)

被引:3
作者
Stehlik, M. [1 ,2 ,3 ]
Kisel'ak, J. [4 ]
Bukina, E. [5 ]
Lu, Y. [6 ]
Baran, S. [7 ]
机构
[1] Johannes Kepler Univ Linz, Dept Appl Stat, AltenbergerStr 69, A-4040 Linz, Austria
[2] Univ Valparaiso, Inst Stat, Valparaiso, Chile
[3] Univ Iowa, Dept Stat & Actuarial Sci, Iowa City, IA 52242 USA
[4] PJ Safarik Univ Kosice, Kosice, Slovakia
[5] Univ Nice Sophia Antipolis, Nice, France
[6] Stanford Univ, Sch Med, Dept Biomed Data Sci, Stanford, CA 94305 USA
[7] Univ Debrecen, Debrecen, Hungary
关键词
FIRCEP procedure; Fredholm integral equation; compound design; estimation; prediction; OPTIMAL DESIGNS; COMPUTER EXPERIMENTS; INFORMATION; PARAMETERS; EQUATIONS;
D O I
10.1080/07362994.2019.1696211
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
We discuss the following problem: Given a set of information criteria for optimal designs, the numerical and computational complexity may drastically differ from one criterion to another. A general novel methodology, called the "FIRCEP" is introduced, and shown to work satisfactorily on a variety of problems relating weighted estimation criterion and Integrated mean square prediction error (IMSPE) prediction criteria in framework of stochastic process. The FIRCEP is shown to be identifying such relationship and providing the exact relations between estimation and prediction for regression problems with correlated errors, without necessity to have known eigenexpansion and truncation methodology. The latter one is the main drawback for automation of complexity reduction algorithms for IMSPE optimization for kernels with unknown eigenexpansion. Thus FIRCEP fills the gap of missing exact method for general kernel satisfying mild regularity conditions in order to develop relation between a class of integrated compound criteria and IMSPE. The exposition proceeds by a series of numerical and real data examples.
引用
收藏
页码:427 / 459
页数:33
相关论文
共 39 条
[1]  
AMARI SI, 1987, DIFFERENTIAL GEOMETR, V10, pI
[3]   Exact optimal designs for computer experiments via Kriging metamodelling [J].
Antognini, Alessandro Baldi ;
Zagoraiou, Maroussa .
JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2010, 140 (09) :2607-2617
[4]   NUMERICAL-SOLUTION OF FREDHOLM INTEGRAL-EQUATIONS OF FIRST KIND [J].
BAKER, CTH ;
WRIGHT, K ;
FOX, L ;
MAYERS, DF .
COMPUTER JOURNAL, 1964, 7 (02) :141-&
[5]   Optimal designs for parameters of shifted Ornstein-Uhlenbeck sheets measured on monotonic sets [J].
Baran, S. ;
Stehlik, M. .
STATISTICS & PROBABILITY LETTERS, 2015, 99 :114-124
[6]   On the optimal designs for the prediction of Ornstein-Uhlenbeck sheets [J].
Baran, Sandor ;
Sikolya, Kinga ;
Stehlik, Milan .
STATISTICS & PROBABILITY LETTERS, 2013, 83 (06) :1580-1587
[7]   Optimal designs for indirect regression [J].
Biedermann, Stefanie ;
Bissantz, Nicolai ;
Dette, Holger ;
Jones, Edmund .
INVERSE PROBLEMS, 2011, 27 (10)
[8]   Design of computer experiments for metamodel generation [J].
Crary, SB .
ANALOG INTEGRATED CIRCUITS AND SIGNAL PROCESSING, 2002, 32 (01) :7-16
[9]   RISK-FACTORS FOR HIP FRACTURE IN WHITE WOMEN [J].
CUMMINGS, SR ;
NEVITT, MC ;
BROWNER, WS ;
STONE, K ;
FOX, KM ;
ENSRUD, KE ;
CAULEY, JC ;
BLACK, D ;
VOGT, TM .
NEW ENGLAND JOURNAL OF MEDICINE, 1995, 332 (12) :767-773
[10]   BONE-DENSITY AT VARIOUS SITES FOR PREDICTION OF HIP-FRACTURES [J].
CUMMINGS, SR ;
BLACK, DM ;
NEVITT, MC ;
BROWNER, W ;
CAULEY, J ;
ENSRUD, K ;
GENANT, HK ;
PALERMO, L ;
SCOTT, J ;
VOGT, TM .
LANCET, 1993, 341 (8837) :72-75