EcoSta 2019: Start Registration
View Submission - EcoSta2019
Title: On local Pearson correlation Authors:  Li-Shan Huang - National Tsing Hua University (Taiwan) [presenting]
Abstract: The classical Pearson correlation measures the linear association between two variables. Its extension to measuring local association is widely used but little is known for its theoretical properties. We investigate the local Pearson correlation for both conditional and symmetric cases and establish properties using a kernel smoothing approach. Simulated examples confirm the asymptotic theory and a real data example is given for illustration.