Bagging-enhanced sampling schedule for functional quadratic regression. (English) Zbl 1478.62154

Summary: Establishing an optimal sampling schedule is a crucial step toward a precise inference of the underlying functional mechanism of a process, especially when data collection is expensive/difficult. This work is concerned with optimal sampling plans for predicting a scalar response using a functional predictor when a quadratic regression relationship is present. An optimality criterion for selecting the best sampling schedules is derived, and some important properties of the criterion are provided. In addition, a bootstrap aggregating (bagging) strategy is proposed to enhance the quality of the obtained sampling schedule.


62H25 Factor analysis and principal components; correspondence analysis
62G08 Nonparametric regression and quantile regression
62H12 Estimation in multivariate analysis
62R10 Functional data analysis
62P10 Applications of statistics to biology and medical sciences; meta analysis


fdapace; PACE; PACE; fda (R)
Full Text: DOI


[1] Bao, Y.; Ullah, A., Expectation of quadratic forms in normal and nonnormal variables with applications, J Stat Plan Inference, 140, 1193-1205 (2010) · Zbl 1181.62075
[2] Breiman, L., Bagging predictors, Mach Learn, 24, 123-140 (1996) · Zbl 0858.68080
[3] Bühlmann, P.; Yu, B., Analyzing bagging, Ann Stat, 30, 927-961 (2002) · Zbl 1029.62037
[4] Bunea, F.; Xiao, L., On the sample covariance matrix estimator of reduced effective rank population matrices, with applications to fPCA, Bernoulli, 21, 1200-1230 (2015) · Zbl 1388.62173
[5] Chen K, Ayutyanont N, Langbaum JB, Fleisher AS, Reschke C, Lee W et al (2011) Characterizing Alzheimer’s disease using a hypometabolic convergence index. Neuroimage 56:52-60
[6] Chen Y, Dai X, Fan J, Hadjipantelis PZ, Han K, Ji H et al (2019) fdapace: functional data analysis and empirical dynamics. R package version 0.5.1
[7] Chernoff, H., Locally optimal designs for estimating parameters, Ann Math Stat, 24, 586-602 (1953) · Zbl 0053.10504
[8] Hsing, T.; Eubank, R., Theoretical foundations of functional data analysis, with an introduction to linear operators (2015), London: Wiley, London · Zbl 1338.62009
[9] James, GM; Hastie, TJ; Sugar, CA, Principal component models for sparse functional data, Biometrika, 87, 587-602 (2000) · Zbl 0962.62056
[10] Ji, H.; Müller, HG, Optimal designs for longitudinal and functional data, J R Stat Soc B, 79, 859-876 (2017) · Zbl 1411.62231
[11] Jiang, CR; Aston, JA; Wang, JL, Smoothing dynamic positron emission tomography time courses using functional principal components, Neuroimage, 47, 184-193 (2009)
[12] Kolibas E, Korinkova V, Novotny V, Vajdickova K, Hunakova D (2000) ADAS-cog (Alzheimer’s disease assessment scale-cognitive subscale) validation of the Slovak version. Eur Psychiatry 15:443
[13] Lai, TL; Shih, MC; Wong, SP, A new approach to modeling covariate effects and individualization in population pharmacokinetics-pharmacodynamics, J Pharmacokinet Pharmacodyn, 33, 49-74 (2006)
[14] Li C (2017) Statistical methods for functional and complex data. North Carolina State University. Ph.D. dissertation
[15] Li, C.; Xiao, L., Statistical methods for functional and complex data. Optimal design for classification of functional data, Can J Stat, 48, 285-307 (2020) · Zbl 1492.62198
[16] Müller, HG; Sen, R.; Stadtmüller, U., Functional data analysis for volatility, J Econom, 165, 233-245 (2011) · Zbl 1441.62817
[17] Park, SY; Xiao, L.; Willbur, JD; Staicu, AM; Jumbe, NL, A joint design for functional data with application to scheduling ultrasound scans, Comput Stat Data Anal, 122, 101-114 (2018) · Zbl 1469.62127
[18] Peng, J.; Paul, D., A geometric approach to maximum likelihood estimation of the functional principal components from sparse longitudinal data, J Comput Graph Stat, 18, 995-1015 (2009)
[19] Ramsay, JO; Ramsay, JB, Functional data analysis of the dynamics of the monthly index of nondurable goods production, J Econom, 107, 327-344 (2002) · Zbl 1051.62118
[20] Ramsay, JO; Silverman, BW, Functional data analysis (2005), New York: Springer, New York · Zbl 1079.62006
[21] Rha, H.; Kao, MH; Pan, R., Design optimal sampling plans for functional regression models, Comput Stat Data Anal, 146, 106925 (2020) · Zbl 07198036
[22] Saleh, M.; Kao, MH; Pan, R., Design D-optimal event-related functional magnetic resonance imaging experiments, J R Stat Soc Ser C, 66, 73-91 (2017)
[23] Sørensen, H.; Goldsmith, J.; Sangalli, LM, An introduction with medical applications to functional data analysis, Stat Med, 32, 5222-5240 (2013)
[24] Tuddenham, RD; Snyder, MM, Physical growth of California boys and girls from birth to age 18, Univ Calif Publ Child Dev, 1, 183-364 (1954)
[25] West, RM; Harris, K.; Giilthorpe, MS; Tolman, C.; Will, EJ, Functional data analysis applied to a randomized controlled clinical trial in hemodialysis patients describes the variability of patient responses in the control of renal anemia, J Am Soc Nephrol, 18, 2371-2376 (2007)
[26] Wu, M.; Diez-Roux, A.; Raghunathan, TE; Sánchez, BN, FPCA-based method to select optimal sampling schedules that capture between-subject variability in longitudinal studies, Biometrics, 74, 229-238 (2018) · Zbl 1415.62143
[27] Xiao, L.; Li, C.; Checkley, W.; Crainiceanu, C., Fast covariance estimation for sparse functional data, Stat Comput, 28, 511-522 (2018) · Zbl 1384.62142
[28] Yao F, Liu B, Tao W, Wu S, Yang NT, Yang W et al (2015) PACE: principal analysis by conditional expectation. MATLAB package version 2.17
[29] Yao, F.; Müller, HG, Functional quadratic regression, Biometrika, 97, 49-64 (2010) · Zbl 1183.62113
[30] Yao, F.; Müller, HG; Wang, JL, Functional data analysis for sparse longitudinal data, J Am Stat Assoc, 100, 577-590 (2005) · Zbl 1117.62451
[31] Yao, F.; Müller, HG; Wang, JL, Functional linear regression analysis for longitudinal data, Ann Stat, 33, 2873-2903 (2005) · Zbl 1084.62096
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.