2003 – 2006
Lucy was Ashish's 4th PhD student, supervised by him and David Nott (in Statistics at UNSW at that time, and an expert on Bayesian methods). She was the first student to work on Bayesian statistical methods, and went on to develop this logic for mixing multiple models as a means of reducing structural uncertainty - called the Hierarchical Mixtures of Experts.
Her PhD consisted of:
- Marshall, L., D. Nott, A. Sharma (2004). "A comparative study of Markov chain Monte Carlo methods for conceptual rainfall-runoff modeling - art. no. W02501." Water Resources Research 40(2): 2501-2501.
- Marshall, L., D. Nott, A. Sharma (2005). "Hydrological model selection: a Bayesian alternative - art. no. W10422." Water Resources Research 41(10): 10422-10422.
- Marshall, L., D. J. Nott, A. Sharma (2007). "Towards dynamic catchment modeling: A Bayesian Hierarchical Mixtures of Experts framework." Hydrological Processes 21: 847-861.
- Marshall, L., A. Sharma, D. Nott. (2006). "Modelling the Catchment via Mixtures: Issues of Model Specification and Validation." Water Resources Research 42(doi:10.1029/2005WR004613): W11409.
- Marshall, L., A. Sharma, D. Nott. (2007). "A single model ensemble versus a dynamic modeling platform: semi-distributed rainfall runoff modeling in a Hierarchical Mixtures of Experts framework." Geophysical Research Letters 34(L01404).
Lucy's PhD led to many later projects and PhDs on the general topic of reducing model structural uncertainty, including the PhDs of Shahadat Chowdhury and Erwin Jeremiah.