In their article, Greven and Scheipl describe an impressively general framework for performing functional regression that builds upon the generalized additive modelling framework. Over the past number of years, my collaborators and I have also been developing a general framework for functional regression, functional mixed models, which shares many similarities with this framework, but has many differences as well. In this discussion, I compare and contrast these two frameworks, to hopefully illuminate characteristics of each, highlighting their respective strengths and weaknesses, and providing recommendations regarding the settings in which each approach might be preferable.
BhattacharyaAPatiDPillaiNSDunsonDB (2015) Dirichlet-Laplace priors for optimal shrinkage. Journal of the American Statis- tical Association, 110, 1479–90.
3.
BrockhausS (2016) FDboost: Boosting fun- ctional regression models. URL http://cran.r-project.org/web/packages/FDboost. R pa- ckage version 0.1-1.
4.
BrockhausSMelcherMLeischFGrevenS (2016) Boosting flexible functional regression models with a high number of functional historical effects. Statistics and Computing. doi:10.1007/s11222-016-9662-1.
5.
BrockhausSScheiplFHothornTGrevenS (2015) The functional linear array model. Statistical Modelling, 15, 279–300.
6.
BuhlmannPHothornT (2007) Boosting algorithms: Regularization, prediction, and model fitting. Statistical Science, 22, 477–505.
7.
CarrollRJRuppertD (1988) Transformation andNew York, NY: Chapman and Hall.
8.
CarvahloCMPolsonNGScottJG (2010) The horseshoe estimator for sparse signals. Biometrika, 97, 465–80.
9.
CederbaumJPouplierMHoolePGrevenS (2015) Functional linear mixed models for irregularly or sparsely sampled data. Statistical Modelling, 16, 67–88.
LeeWBaladandayuthapaniVFazioMDownsCMorrisJS (2016) Semiparametric functional mixed models for longitudinally correlated functional data with application to glaucoma data. Under revision.
18.
LeeWMorrisJS (2015) Identification of differentially expressed methylated loci using wavelet-based functional mixed models. Bioinformatics, 32, 664–72.
19.
MalloyEJMorrisJSAdarSDSuhHHGoldDRCoullBA (2010) Wavelet-based functional linear mixed models: An application to measurement error connected distributed lag models. Biostatistics, 11, 432–52.
20.
MartinezJGBohnKMCarrollRJMorrisJS (2013) A study of Mexican free-tailed bat syllables: Bayesian functional mixed modelling of nonstationary acoustic time series. Journal of the American Statistical Association, 108, 514–26.
21.
McLeanMWHookerGStaicuA-MScheiplFRuppertD (2014) Functional generalized additive models. Journal of Computational and Graphical Statistics, 23, 249–69.
MeyerMMorrisJSMorrowJDHershCPLangeCCoullBA (2016) Ordinal probit wavelet-based functional models for eQTL analysis. Under revision.
24.
MorrisJS (2013) Statistical methods for proteomic biomarker discovery using feature extraction or functional data analysis approaches. Statistics and its Interface, 5, 117–136.
25.
MorrisJS (2015) Functional regression. Annual Review of Statistics and Its Application, 2, 321–59.
26.
MorrisJSArroyoCCoullBARyanLMHerrickRCGortmakerSL (2006) Using wavelet-based functional mixed models to characterize population heterogeneity in accelerometer profiles: A case study. Journal of the American Statistical Association, 101, 1352–64.
27.
MorrisJSBaladandayuthapaniVHerrickRCSannaPPGutsteinH (2011) Automated analysis of quantitative image data using isomorphic functional mixed models, with application to proteomics data. Annals of Applied Statistics, 5, 894–923.
28.
MorrisJSBrownPJHerrickRCBaggerlyKACoombesKR (2008) Bayesian analysis of mass spectrometry data using wavelet-based functional mixed models. Biometrics, 12, 479–89.
29.
MorrisJSCarrollRJ (2006) Wavelet-based functional mixed models. Journal of the Royal Statistical Society, 68, 179–99.
30.
MorrisJSVannucciMBrownPJCarrollRJ (2003) Wavelet-based nonparametric modelling of hierarchical functions in colon carcinogenesis. Journal of the American Statistical Association, 98, 573–83.
31.
ParkTCasellaG (2008) The Bayesian lasso. Journal of the American Statistical Association, 103, 681–86.
32.
RamsayJOSilvermanBW (2005) Functional Data Analysis, 2nd edition. New York: Springer-Verlag.
33.
RauschPMorrisJSSommerWKrifkaM (2013) When you are thrown a curve: Two R packages for swerving with wavelet-based functional mixed models. Proceedings of the Linguistic Evidence Conference, Berlin, Germany. Available at https://www2.hu-berlin.de/linguistic-evidence-berlin-2013/[download/abstracts/Rausch_Morris_][Sommer_Krifka-colored_LinguisticEvidence][2013.pdf]
34.
RuppertDWandMPCarrollRJ (2003) Semiparametric Regression.New York, NY: Cambridge University Press.
ScheiplFGrevenS (2016) Identifiability in penalized function-on-function regression models. Electronic Journal of Statistics, 10, 495–526.
37.
ScheiplFStaicuAMGrevenS (2015) Functional additive mixed models. Journal of Computational and Graphical Statistics, 24, 477–501.
38.
WoodSN (2016) mgcv: Mixed GAM computatio- nal vehicle with GCV/AIC/REML smoothness estimation. URL http://CRAN.R-proj[ect.org/package-mgcv.] R package version 1.8-12.
39.
ZhangLBaladandayuthapaniVZhuHBaggerlyKAMajewskiTACzerniakBAMorrisJS (2016a) Functional CAR models for large spatially correlated functional datasets. Journal of the American Statistical Association, 111, 772–86.
40.
ZhangLBaladandayuthapaniVVersaceFCinciripiniPMorrisJS (2016b) Bayesian functional graphical modelling. Under revision.
41.
ZhuHBrownPJMorrisJS (2011) Robust, adaptive functional regression in functional mixed model framework. Journal of the American Statistical Association, 106, 1167–79.
42.
ZhuHBrownPJMorrisJS (2012) Robust classification of functional and quantitative image data using functional mixed models. Biometrics, 68, 1260–68.
43.
ZhuHCaspersPMorrisJSWuXMullerR (2016a) A unified analysis of sonar-terrain data using Bayesian functional mixed models. Technometrics, in press.
44.
ZhuHMorrisJSWeiFCoxDD (2016b) Multivariate functional response regression, with application to fluorescence spectroscopy in a cervical pre-cancer study. Under revision.
45.
ZhuHVersaceFCinciripiniPMorrisJS (2016c) Robust functional mixed models for spatially correlated functional regression, with application to event-related potentials for nicotine-addicted individuals. Under revision.
46.
ZipunnikinovVCaffoBYousemDDavitzikosCSchwartzCrainiceanuC (2011) Multilevel functional principal component analysis for high-dimensional data. Journal of Computational and Graphical Statistics, 20, 852–73.
