About

Start your brilliant career with a degree from Australia's #1 ranked university.

News

Find contact details for any general enquiries.

Engage with us

Find contact details for any general enquiries.

Study with us

Our graduates gain the knowledge and skills to lead organisations, develop public policy, create new companies and undertake research.

Our research

Our research is focused on issues that are highly significant for organisations, the Australian economy, and society at large.

Current Students

Whether you're a new or continuing student, you can find everything you need here about managing your program and the opportunities available to you.

Alumni

Our alumni may be found in the world’s leading companies, policy agencies and universities.

Contact us

Find contact details for any general enquiries.

RSFAS Seminar | Dr Jia Liu

RSFAS Seminar | Dr Jia Liu

Model based Bayesian spatio-temporal survey design for species distribution modelling.

Available to CBE staff and HDRs only.

In geostatistics, the spatiotemporal design for data collection is central for accurate prediction and parameter inference. An important class of geostatistical models is log-Gaussian Cox process (LGCP) but there are no formal analyses on spatial or spatiotemporal survey designs for them. In this talk, I will first go through the traditional balanced and uniform random designs in situations where analyst has prior information on intensity function of LGCP and show that these designs are inefficient in such situations. I will introduce a new design sampling method named a rejection sampling design, which extends the traditional balanced and random designs by directing survey sites to locations that are a priori expected to provide most information. Comparison between the new method and the traditional designs will be demonstrated. I will give an example of a case study, which concerns planning a survey design for analyzing larval areas of two commercially important fish stocks on Finnish coastal region. Our experiments show that the designs generated by the proposed rejection sampling method clearly outperform the traditional balanced and uniform random survey designs. Moreover, the method is easily applicable to other models in general. To extend the topic, I will also talk about Bayesian optimal design and large matrix calculation.

Updated:   4 May 2020 / Responsible Officer:  CBE Communications and Outreach / Page Contact:  College Web Team