Paul S.P. Cowpertwait
Associate Professor - Mathematical Sciences (Analytics)
School of Computing and Mathematical Sciences
Auckland University of Technology
AUT Tower, 2-14 Wakefield StreetEmail: paul.cowpertwait@aut.ac.nz
|
|
Paul S.P. Cowpertwait Associate Professor - Mathematical Sciences (Analytics) School of Computing and Mathematical Sciences Auckland University of Technology AUT Tower, 2-14 Wakefield StreetEmail: paul.cowpertwait@aut.ac.nz |
|
| Contents | ||
|---|---|---|
| Analytics | Research Interests and Projects | |
| External Research and Consultancy Contracts | Some Publications | |
In general terms, analytics means methods of logical analysis. However, in applications, analytics has come to mean the use of computer technology and statistics to solve problems in business and industry so that, in this sense, analytics is strongly linked to information systems and software environments. To reflect this analytics at AUT has a strong foundation in mathematics, logic and programming, followed by more vocational subjects in the latter part of the programme. A suitable and popular choice is to do a double major in Analytics with either Computer Science or Mathematics. Using suitable electives students may also choose to do a minor in information systems.
The analytics major at AUT consists primarily of three streams: (i) applied statistics; (ii) data mining and programming; (iii) applied probability and stochastic modelling. Papers that form the core to the analytics major include: Probability, Applied Statistics, Functional Programming, Logical Database Design, Statistical Methods, Statistical Theory, Forecasting, Operations Research, Data Mining, Data Management using SAS, Artificial Intelligence, Industrial and Business Analytics, and Random Systems. The ideology underpinning the S language and, especially, R (being an open source environment with properties of a functional language), also forms a thread through the major. Although proprietary software is required for courses that have particular industrial and business applications, where possible, the selected software environments are open source and free to foster the sharing of knowledge with the wider community.
With the advancement of computing technology and statistical methodology, opportunities for employment in the field of analytics has grown considerably (e.g. see the resources at KDnuggets).
Stochastic rainfall and temperature models for the Basque Country, Spain. Development of full spatial temporal stochastic rainfall and temperature models for use in hydrological catchment models; developing a suitable regionalization procedure.
Development of a fine resolution stochastic model of rainfall based on point processes (the Bartlett-Lewis Pulse model). (In collaboration with Valerie Isham and Christian Onof.)
Research in using R for time series and stochastic processes (with Andrew Metcalfe, University of Adelaide).
Partner investigator on Australian Research Council (ARC) grant. The project aims to develop a spatially consistent stochastic rainfall model (based on the Neyman-Scott process) for simulation of hourly series which can be used for flow simulation studies and drought analysis for large regions in Australia. (Collaboration with Prof Martin Lambert, A/Prof Andrew Metcalfe, and Michael Leonard, University of Adelaide.)
Development of fitting procedures for fine scale stochastic models. Engle's ACD model for financial series and the BLP model for rainfall (with John Xie and Danny Walsh).
Spatial-temporal rainfall simulation. The project aim is to develop a full spatial temporal rainfall model, based on the Neyman-Scott Rectangular Pulses model, for Malaysia. (with Barry McDonald and Azlina Ismail).
Research in statistical programming (with Bruce Mills).
Water Research Centre (WRc, UK) and Acea SpA (Italy), 2010. Development of a spatial-temporal point process rainfall model for the Roma region.
Scottish Water PLC (2008). Consultant to the Water Research Centre (WRc, UK) acting on behalf of Scottish Water. Project description: Simulation of multisite hourly rainfall data over the Glasgow catchment, Scotland; infilling missing values, generating data at sites with no available historical series, generation of long multisite series for the purpose of hydraulic urban drainage design and minimising the flood risks associated with heavy rainfall. For use in a six-million pound project coordinated by the Metropolitan Glasgow Strategic Drainage Partnership.
Scottish Water PLC (2007). Consultant to the WRc (UK) acting on behalf of Scottish Water. Simulation of multisite hourly rainfall data over the Irvine catchment, Scotland, allowing for orographic effects due to altitude; simulation of long spatial-temporal series for the purpose of urban drainage design and analysis. For use in hydrological studies aimed at improving water quality in the Irvine catchment; some details given here.
Thames Water PLC, UK (2005-2006). Consultant to the WRc (UK) acting on behalf of Thames Water, London. Project description: Full spatial-temporal rainfall data simulation over London, allowing for orographic effects due to altitude; simulation of long spatial-temporal series for the purpose of urban drainage design and hydraulic systems analysis. For use in the Thames Tideway Tunnels project. Further details of the rainfall model can be found here.
Metrowater, Auckland City (2002-2006). Project description: Generation of long records of rainfall time series (500 years) for the current climate and a future climate characterised by enhanced concentrations of greenhouse gases. This was required for the Integrated Catchment Study (ICS) of Auckland's drainage network, a NZ$23.5 million project initiated by Metrowater in consultation with Auckland Regional Council. Further details can be found here.
North Shore City Council (2003). Consultant to North Shore City Council. Project description: Extending current rainfall records for the design and upgrading of the North Shore's drainage system; simulation of future rainfall scenarios to prepare North Shore City for a change in climate. Some details of the overall project can be found here.
Water Research Centre, UK (2000). Development of algorithms to improve the simulation of extreme rainfall events for urban drainage applications. Algorithms implemented into the UK rainfall simulation package (Stormpac).
Cowpertwait P. S. P. (2010). A Neyman-Scott model with continuous distributions of storm types, Australian and New Zealand Industrial and Applied Mathematics Journal, 51, 97-108.
Cowpertwait P.S.P., Salinger J., and Mullan B. (2009). A Spatial-temporal Stochastic Rainfall Model for Auckland City: Scenarios for Current and Future Climates. Journal of Hydrology (New Zealand), Vol. 48, No. 2: 95-109. here.
Cowpertwait P.S.P. and Metcalfe A.V. (2009). Introductory Time Series with R. Springer. For details at Springer click here. A review of the book can be found here.
Leonard M., Lambert M.F., Metcalfe A.V., and Cowpertwait P.S.P. (2008). A space-time Neyman-Scott rainfall model with defined storm extent. Water Resources Research . For more information click here.
Burton A, Kilsby C.G., Fowler H.J., Cowpertwait P.S.P., and O'Connell P.E. (2008). RainSim: A spatial temporal stochastic rainfall modelling system. Environmental Modelling and Software. For more information click here.
Cowpertwait P., Isham V., and Onof C. (2007). Point process models of rainfall: Developments for fine-scale structure. Proceedings of the Royal Society of London, Series A. For more information click here.
Cowpertwait P. (2006). A spatial-temporal point process model for the Thames catchment, London. Journal of Hydrology. For more information click here.
I have been involved in the development of the WRc (Water Research Centre) rainfall time series simulation software Stormpac. This is the UK's leading software for rainfall simulation for urban catchment studies and hydraulic systems analysis. For more information refer to the User Guide.
I have a range of up-to-date rainfall simulation software which may be available for consultancy work upon request.