Multi-aspect local inference for functional data: Analysis of ultrasound tongue profiles

JOURNAL OF MULTIVARIATE ANALYSIS

卷:170

Article

Pini, Alessia^1,2  Spreafico, Lorenzo³  Vantini, Simone⁴  Vietti, Alessandro⁵ 

[1] Umea Univ, Umea Sch Business Econ & Stat, Dept Stat, SE-90187 Umea, Sweden

[2] Univ Cattolica Sacro Cuore, Dept Stat Sci, Largo A Gemelli 1, I-20123 Milan, Italy

[3] Austrian Acad Sci, Acoust Res Inst, Vienna, Austria

[4] Politecn Milan, Dept Math, MOX, Pza Leonardo da Vinci 32, I-20133 Milan, Italy

[5] Free Univ Bozen Bolzano, ALPS, Bolzano, Italy

关键词: Articulatory phonetics;    Derivatives;    Functional data analysis;    Inference;    Interval-wise error rate;   

DOI  :  10.1016/j.jmva.2018.11.006

来源: Elsevier

【 摘 要 】

Motivated by the analysis of a dataset of ultrasound tongue profiles, we present multi-aspect interval-wise testing (IWT), i.e., a local nonparametric inferential technique for functional data embedded in Sobolev spaces. Multi-aspect IWT is a nonparametric procedure that tests differences between groups of functional data, jointly taking into account the curves and their derivatives. Multi-aspect IWT provides adjusted multi-aspect p-value functions that can be used to select intervals of the domain that are imputable for the rejection of a null hypothesis. As a result, it can impute the rejection of a functional null hypothesis to specific intervals of the domain and to specific orders of differentiation. We show that the multi-aspect p-value functions are provided with a control of the family wise error rate and that they are consistent. We apply multi-aspect IWT to the analysis of a dataset of tongue profiles recorded for a study on Tyrolean, a German dialect spoken in South Tyrol. We test differences between five different ways of articulating the uvular /r/: vocalized /r/, approximant, fricative, tap, and trill. Multi-aspect IWT-based comparisons result in an informative and detailed representation of the regions of the tongue where a significant difference occurs. (C) 2018 Elsevier Inc. All rights reserved.

【 授权许可】

Free