Selection of research projects in Theoretical and Applied Statistics
Smoothing and Regression in Palaeo research: species distributions, fire, climate and long-term variations in vegetation in Switzerland
(Funding: Swiss National Science Foundation, Division of Mathematics, Natural & Engineering Sciences)In palaeoenvironmental research, stable isotope ratios of oxygen are used as temperature proxies and have been proven to be useful for quantifying
temperature changes of the past. Similarly, charcoal records serve as proxies for fire events whereas pollen assemblages are used to assess presence of tree species in the region. One issue is that strong fluctuations in the
environmental conditions may lead to plant species becoming extinct or abundant.
Thus the range of variability in the vegetation response over time, as well as how
this has been influenced by fluctuations in the environmental conditions in the past are of considerable interest.
Collaborators: Patricia Menendez(WSL; Ph.D. candidate, ETH:2005-2007)
Institute for Plant Sciences, University of Bern, Willy Tinner & Brigitta Ammann
University of Bergen, John Birks
Seminar für Statistik, ETH, Zürich, Hans Künsch
Some publications:
2006 [submitted] Menendez, P. & Ghosh, S. On some nonparametric smoothing methods for assessing climate change'
Proceedings of the Joint Statistical Meetings (JSM), August
6 - 10, 2006
Quantile estimation for time series with applications to the Swiss precipitation records.
(Funding:Swiss National Science Foundation, 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. This aim of this project is to develop some nonparametric estimation & prediction methods for assessing changing precipitation patterns in Switzerland.
Collaborators:
Dana Draghicescu (WSL:1999-2002. Doctoral thesis 2002, EPFL; now at CUNY, USA)
Institute for Climate Research, ETH, Christoph Frei; now at Meteoswiss, Zurich
Department of Mathematics, Applied Statistics Group, EPFL, Lausanne, Stephan Morgenthaler Some publications:
2001 Ghosh, S., Draghicescu, D. Quantile estimation to assess extreme climate events. Geophysical Research Abstracts, Volume 3, 2001, European Geophysical Society, Nice, France, 25-30 March, 2001.
2002 Ghosh, S., Draghicescu, D. An algorithm for optimal bandwidth selection for smooth nonparametric quantiles and distribution functions. In, Statistics in Industry and Technology: Statistical Data Analysis based on the L1-norm and related methods, Birkhäuser Verlag, Basel, Switzerland, pp. 161-168.
2002 Ghosh, S., Draghicescu, D. Predicting the distribution function for long-memory processes. International Journal of Forecasting 18: 283-290.
2002 Draghicescu, D. Nonparametric quantile estimation for depepdent data. Ph.D. thesis, EPFL.
2003 Draghicescu, D., Ghosh, S. Smooth nonparametric quantiles. In, Proceedings of the 2nd International colloquium of Mathematics in Engineering and Numerical Physics (MENP-2), Geometry Balkan Press, Bucharest, Romania 2003, pp. 45-52.
Time series methods with environmental and other applications.
Ongoing research among statisticians, on statistical methods for time series data. The topics include developing models and tests that would distinguish between deterministic
and spurious trends, data driven methods, as well as merging of physical models with empirical evidences so as to bridge gaps between theory and data.
Collaborators: University of Washington, Seattle Department of Mathematics & Statistics, Konstanz Heriot-Watt University Some publications:
2000 Beran, J., Ghosh, S. Estimation of the dominating frequency for stationary and non-stationary fractional autoregressive models. Journal of Time Series Analysis, Vol. 21, No. 5, 517-533.
2001 Ghosh, S. Nonparametric trend estimation in replicated time series. Journal of Statistical Planning and Inference, Vol. 97, 263-274.
2002 Beran, J., Feng, Y., Ghosh, S, Sibbertsen, P. On robust local polynomial estimation with long-memory errors. International Journal of Forecasting 18: 227-241
2003 Beran, J., Ghosh, S, Sibbertsen, P. Nonparametric M-estimation with long-memory errors. Journal of Statistical Planning and Inference 117: 199-205.
2007 Ghosh, S., Beran, J., Heiler, S., Percival, D., Tinner, W. Analysis of environmental time series. (book chapter submitted)
****[In Preparation, Text book] Beran, J., Feng, Y., Ghosh, S. Nonparametric and Semiparametric estimation methods for time series.