Geospatial analysis of multivariate processes applied to soil pollution data

Majumdar’s work studies geospatial processes, especially the covariance structure of multivariate geospatial variables that cannot be explained or modeled using usual stationary structures. It extends parametric methods to a parsimonious semiparametric model using a kernel convolution technique. This work has been applied to soil pollution data in a heterogeneous ecology in Phoenix, Arizona. Geospatial statistics and analysis have been widely applied in biostatistics and genomics, and to study outbreaks of disease in epidemiological studies.

Email [email protected] for the Zoom link.