1,721,548 research outputs found
The Sound of Silence: Listening to the Villagers to Learn about Orangutans
No full textAbstract
Orangutans, those shy and gentle primates of Borneo and Sumatra, are iconic of species under threat. Efforts to conserve them have met with little success. But Kerrie Mengersen and her colleagues have brought statistics to bear to provide practical guidance towards saving the creature whose name means “old man of the forests”.</jats:p
Heritability and Linkage Analysis of Appendicitis Utilizing Age at Onset
Get PDFAppendicitis usually afflicts the young, but there is a large tail in the distribution of onset age. The genetics of this disease are still not well understood. A heritability analysis and genome wide linkage analysis of a large twin dataset was undertaken. Treating age of onset of appendicitis as a censored survival trait revealed a heritability of 0.21, and found evidence of linkage to Chromosome 1p37.3. Author(s): Christopher Oldmeadow 1 * | Kerrie Mengersen 2 | Nicholas Martin 3 | David L. Duffy
Supplemental Material - Patterns of educational performance among Indigenous students in Australia, 2010–2019: Within-cohort, peer matching analysis for data-led decision-making
No full textSupplemental Material for Patterns of educational performance among indigenous students in Australia, 2010–2019: Within-cohort, peer matching analysis for data-led decision-making by Peter J. Anderson, Owen Forbes, Kerrie Mengersen, and Zane M. Diamond in Australian Journal of Education</p
larc - Least Angle Regression Companion
No full text<p>This repository contains the data and code necessary to replicate the analysis described in the PLOS ONE article:by , David W. Lamb (DWL) and Kerrie Mengersen (KM).</p>
<p>Code and repository authorship was the sole responsibility of Benjamin R. Fitzpatrick.</p>
<p>The code file <code>example_analysis.R</code> illustrates how the functions included in this repository may be used to replicate the analysis described in the article. The article discusses the relevant theory and demonstrates the application of these methods to a geostatistical case study. This repository contains a set of functions written in the . The analysis this repository enables makes heavy use of the Least Angle Regression (LAR) algorithm for finding Least Absolute Shrinkage Selection Operator (LASSO) regularised solutions to multiple linear regression problems. An R package for conducting Least Absolute Shrinkage Selection Operator (LASSO) variable selection with the LAR algorithm already exists and is hosted on the Comprehensive R Archive Network under the name ''. This repository makes heavy use of functions from the 'lars' package.</p>
<p>This repository contains functions that:</p>
<ul>
<li>randomly generate unique divisions of a sequence of numbers into two groups of user specified sizes (the intent being that these two groups of numbers are used as row indices to create training and validation sets from a full dataframe)</li>
<li>use the LAR algorithm within a cross validation scheme in a manner that permits greater control of the particulars than is provided by the <code>cv.lars( )</code> function from the 'lars' package</li>
<li>use chord diagrams to visualise the covariate selection frequencies that result from conducting LAR within a cross validation scheme</li>
<li>model average the predictions from the models selected for each of the training sets in the cross validation scheme</li>
<li>interpolate a geostatistical response variable to a full cover predicted raster via such model averaged predictions.</li>
</ul>
<p>The functions provided here depend on the R packages:</p>
<ul>
<li></li>
<li></li>
<li></li>
<li></li>
<li></li>
</ul>
Code and selected model outputs for "Analysis of sloppiness in model simulations: unveiling parameter uncertainty when mathematical models are fitted to data"
No full textAll of the code, and selected model output files, associated with the paper "Analysis of sloppiness in model simulations: unveiling parameter uncertainty when mathematical models are fitted to data ".The code and model output files contained in this repository are sufficient to generate all results shown in the paper (including all of the results shown in Supporting Information).Authors: Gloria M. Monsalve Bravo, Brodie A. J. Lawson, Christopher Drovandi, Kevin Burrage, Kevin S. Brown, Christopher M. Baker, Sarah A. Vollert, Kerrie Mengersen, Eve McDonald-Madden, and Matthew P. Adams</p
Estimating detection rates and probabilities
Full text linkQuantitative Approaches Frith Jarrad, Samantha Low-Choy, Kerrie Mengersen ...
CAB International 2015. Biosecurity Surveillance: Quantitative Approaches (eds
1 Introduction to Biosecurity Surveillance: Quantitative Approaches
Mr Benjamin Fitzpatrick
No full text<p>Ben Fitzpatrick's research focuses on quantifying soil carbon stocks at an individual paddock scale. He is exploring methods to model small numbers of geostatistical observations of soil carbon with broad collections of environmental data all available at much finer spatial resolutions than the soil carbon data. The aim of this modelling is to use these high resolution environmental data to assist the interpolation of the soil carbon observations to full cover maps. He is also interested in visualisation methods to communicate predictions from models, the uncertainty associated with these predictions and the mechanics of the models producing these predictions.. His research project has been awarded a Postgraduate Scholarship from the CRC for Spatial Information (CRCSI).</p>
Barrow Island's biosecurity : catching the unknown invader
No full textWhen industry meets a conservation area, animals or plants from outside may hitch a lift and potentially wreak havoc. How can you be sure of catching the intruders – or at least 80% sure? A government directive instructed Frith Jarrad, Peter Whittle, Susan Barrett and Kerrie Mengersen to come up with a statistically measurable scheme
Data derived state probabilities for Z. noltei monitoring study. Observed variables were shoot density at four sites in this study
No full text<p>Data-derived state probabilities for the seagrass monitoring study (Cognat et al., 2018) were used to validate the DBN model for<em> Zostera noltei</em> in Arcachon Bay. Shoot density was observed at four sites in this study.</p>
- …
