Test of Significance When Data are Curves

摘要: Abstract With modern technology, massive data can easily be collected in a form of multiple sets curves. New statistical challenge includes testing whether there is any statistically significant difference among these In this article we propose some new tests for comparing two groups curves based on the adaptive Neyman test and wavelet thresholding techniques introduced earlier by Fan. We demonstrate that inherit properties outlined Fan they are simple powerful detecting differences between then further generalize idea to compare curves, resulting an high-dimensional analysis variance, called HANOVA. These newly developed illustrated using dataset pizza commercials where observations cornea topography ophthalmology images individuals observed. A simulation example also presented illustrate power adapti...