Collaborative research projects in Theoretical and Applied Statistics

Contact Sucharita Ghosh (Email: rita.ghosh@wsl.ch)


  • Quantile estimation for time series with applications to the Swiss precipitation records (supported by the SNF - Division of Mathematics, Natural & Engineering Sciences) Due to the adverse impacts of extreme weather phenomena in practically all spheres of life, there is an increasing interest in modeling long term stochastic variations in climate events. As regards heavy precipitation, recent studies indicate that the frequencies of such events may have increased in the Alpine region requiring a proper understanding of such phenomena through estimation and prediction of the probability distributions of precipitation events. WSL is involved in a joint collaboration with the Institute of Climate Research (ETHZ) and the Department of Mathematics (EPFL) to develop statistical methods to assess the changing precipitation patterns in Switzerland.

      Collaborators:
    1. Ph.D. student: Dana Draghicescu
    2. Institute for Climate Research, ETH, Zürich Christoph Frei
    3. Department of Mathematics, EPFL, Lausanne Stephan Morgenthaler

  • Modeling assimilation activity using Chlorophyll a fluorescence: Diagnosis & Prediction
    The aim of this project is to model typical behavior as opposed to behavior under stress of plants via analysis of chlorophyll a fluorescence curves. Responses of different plant species under varying treatment conditions will be examined in the light of their photosynthetic activities. The statistical methods to be used here include time series analysis of repeated measures data. In paricular, fluorescence measurements on leaves will be taken for two seconds giving rise to some 1200 observations. Typically, the fluorescence curves exhibit a fast rise to a maximum followed by a slow decline to a steady state. One of the specific aims here will be to estimate the theoretical (deterministic trend) fluorescence curve as a function of time and examine how its shape may change depending on the gradual effect of heavy metals and acid rains.

      Collaborators:
    1. University of GenevaThe Bioenergetics Lab
    2. Swiss Federal Research Institute WSL Research Group Bioindications
    3. University of KonstanzDepartment of Mathematics & Statistics
    4. Swiss Federal Research Institute WSLVon der Zelle zum Baum

  • Tree ring web and alternative chronologies
    Construction of alternatives chronologies involves development of appropriate statistical models that enable assessment of variabilities as well as quantiles (in particular extreme quantiles) of the various synchronized tree ring parameters. Selected alternative chronologies will be made available on the Tree Ring Web in addition to the standard (mean) chronologies. This research has among others, implications for research in climate change.

      Collaborators:
    1. Swiss Federal Research Institute WSLDirk Schmatz
    2. Swiss Federal Research Institute WSLIris Heller


  • Statistics of tree rings - Theory, Software, Web
    New time series approaches will be used to assess the long-term behavior of the various tree ring parameters. Special attention will be given to (a) estimation of dominating frequencies ("is there a pulse?"), (b) alpha-quantile chronologies (the applied aspects for this topic will be handled in the previous project, whereas the theoretical model building will be done within this project), (c) the sample size issue revisited ("how much more do we learn by taking additional samples?"), (d) correlations with climate change (partial linear models and related approaches).

      Collaborators:
    1. Swiss Federal Research Institute WSLDirk Schmatz
    2. Swiss Federal Research Institute WSLIris Heller
    3. Swiss Federal Research Institute WSLChristian Hoffmann


  • Time series methods - applications in ecology, finance, etc.
    Building time series models that would distinguish between deterministic as opposed to spurious (due to correlations) trends is important for many environmental problems. Another concern is to merge physical models with empirical evidences so as to bridge gaps between theory and data. Some of the fields of applications for this researh include environment (studies of the polar ice thickness, global warming, local and global climate variations, plant physiology), finance (stock market assessments) and health.

      Collaborators:
    1. University of Washington, Seattle
    2. Department of Mathematics & Statistics, Konstanz
    3. Department of Statistics, Dortmund



  • Rita Ghosh