... Ryden: Inference in Hidden Markov Models Chan/Tong: Chaos: A Statistical Perspective Chen/Shao/Ibrahim: Monte Carlo ... of Populations Hall: The Bootstrap and Edgeworth Expansion Hdrdle: Smoothing Techniques: With Implementation in S... more
... Ryden: Inference in Hidden Markov Models Chan/Tong: Chaos: A Statistical Perspective Chen/Shao/Ibrahim: Monte Carlo ... of Populations Hall: The Bootstrap and Edgeworth Expansion Hdrdle: Smoothing Techniques: With Implementation in S Harrell: Regression Modeling ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests: Zoology, Biology, Ecology, Population ecology, Hymenoptera, and 4 moreParasitoid, Mathematical Model, Musca, and Pupa
Research Interests: Biology and Microarray
ABSTRACT Various models that have been proposed to describe the avoidance of super-parasitism in single ovipositing parasite-host systems are surveyed and presented within a framework that clarifies the assumptions. Computations on the... more
ABSTRACT Various models that have been proposed to describe the avoidance of super-parasitism in single ovipositing parasite-host systems are surveyed and presented within a framework that clarifies the assumptions. Computations on the information statistics are summarized, and a new procedure, which is a variant of a sequential occupancy model, is illustrated on data for wasps laying eggs in some housefly pupae; there are also comparative analyses with some older methods. All models can be formulated within the context of the time-dependent behaviour of a Markov process. Within this context inferences concerning the parameter(s) measuring any avoidance of superparasitism can readily be made conditional on the observed number of ovipositions, thus avoiding estimation of the expected number of ovipositions as a nuisance parameter.
Research Interests:
... of Data Collection _____19 Decision Trees, Artificial Neural Networks (Neural Nets ... for managing large information repositories and thematic information retrieval. ... There is one further program Digital Media Libraries. ...
Research Interests:
Research Interests:
The temperature sensitivity of three egg ages of Plodia interpunctella (Hübner) were examined at 42–48°C heat (hot water) only treatments, 0.5–11.5°C cold storage only treatments, or combination treatments of 42 or 46°C, followed by... more
The temperature sensitivity of three egg ages of Plodia interpunctella (Hübner) were examined at 42–48°C heat (hot water) only treatments, 0.5–11.5°C cold storage only treatments, or combination treatments of 42 or 46°C, followed by subsequent 0–10.5°C cold storage. The 1-day-old egg was typically more tolerant of heat only treatments than the 2- and 3-day-old egg. A 99% mortality of all
Research Interests:
Research Interests:
Research Interests:
Abstract In small plot field trials for insecticidal control of artificial infestations of greasy cutworm (Agrotis ipsilon Hufnagel) in seedling maize, deltamethrin, permethrin, alphamethrin, fenvalerate, and flucythrinate sprayed at... more
Abstract In small plot field trials for insecticidal control of artificial infestations of greasy cutworm (Agrotis ipsilon Hufnagel) in seedling maize, deltamethrin, permethrin, alphamethrin, fenvalerate, and flucythrinate sprayed at recommended rates, prevented plant damage more effectively than endosulfan, chlorpyrifos, and diazinon. All of the above pyrethroid insecticides adequately controlled cutworm infestations at recommended rates of application. The analysis required a generalisation of logit analysis in which both control survival of plants and the treatment effect depend on design factors.
Research Interests:
Research Interests:
Languages shape the way we think, and determine what we can think about. [Benjamin Whorf.] S has forever altered the way people analyze, visualize, and manipulate data... S is an elegant, widely accepted, and enduring software system,... more
Languages shape the way we think, and determine what we can think about. [Benjamin Whorf.] S has forever altered the way people analyze, visualize, and manipulate data... S is an elegant, widely accepted, and enduring software system, with conceptual integrity, thanks to the insight, taste, and effort of John Chambers. [From the citation for the 1998 Association for Computing Machinery Software award.] 2
Research Interests:
Research article Microarray analysis identifies candidate genes for key roles in coral development
Title Functions and data for a course on modern regression and classification
Research Interests:
LazyData yes Description Functions for demographic analysis including lifetable calculations; Lee-Carter modelling; functional data analysis of mortality rates, fertility rates, net migration numbers; and stochastic population... more
LazyData yes Description Functions for demographic analysis including lifetable calculations; Lee-Carter modelling; functional data analysis of mortality rates, fertility rates, net migration numbers; and stochastic population forecasting. License GPL (> = 2)
Research Interests:
Research Interests:
of Variable Selection John Maindonald October 7, 2013 The function varselect() in the leaps package can be used for variable selection. Available appraoches are forward, backward and exhaustive selection. The DAAG package has the... more
of Variable Selection John Maindonald October 7, 2013 The function varselect() in the leaps package can be used for variable selection. Available appraoches are forward, backward and exhaustive selection. The DAAG package has the functions bestsetNoise() and bsnVaryNvar() that are designed to give insight on the sampling properties of output from the function lm(), when one of these variable selection approaches has been used to choose the explanatory variables that appear in the model.
Research Interests:
Research Interests:
Title Functions and data for a course on modern regression and classification Version 0.52
Research Interests:
Research Interests:
Title Use known groups in high-dimensional data to derive scores for plots
These consolidate the earlier separate sets of additional notes – I & II
Research Interests:
Description Data sets and functions, for the display of biological expression array (microarray) data, and for demonstrations with such data LazyLoad true LazyData true
Research Interests:
and data for a course on modern regression and classification
Research Interests:
and data for a course on modern regression and classification
Research Interests:
known groups in high-dimensional data to derive scores for plots
The example data will be for spotted (two-channel) microarrays. Exactly the same approaches are relevant to spotted oligonucleotides. 1 Spotted Microarray Methodology & Tools Each array (“slide”) typically compares expression in genes... more
The example data will be for spotted (two-channel) microarrays. Exactly the same approaches are relevant to spotted oligonucleotides. 1 Spotted Microarray Methodology & Tools Each array (“slide”) typically compares expression in genes from one “sample ” (type of cell) with expression in genes from another “sample”. The common dyes are Cy5 (“red”) and Cy3 (“green”). Steps in the process of getting microarray intensity measurements are: 1. cDNA samples are obtained by reverse transcription, one for each of the two mRNA samples. One sample is labeled with “red ” dye and one with “green ” dye. 2. The samples are mixed and hybridized onto the slide. 3. An image analyser is used to give separate images of the “red ” and “green ” signals. Think of these as much like digital camera images, but with just two frequencies. The resolution may be 15-30 pixels per spot. 4. Image analysis software is used to extract various statistical summaries of the images of the spots and of their surroundings...
The opening and dedication of the new Unitarian Church, Ponsonby Road, took place on Wednesday. The Rev. W. Jellie presided at the dedication service. The church was well filled, and the walls were tastefully decorated. The choir and... more
The opening and dedication of the new Unitarian Church, Ponsonby Road, took place on Wednesday. The Rev. W. Jellie presided at the dedication service. The church was well filled, and the walls were tastefully decorated. The choir and orchestra were under the baton of Mr. Gribble, choirmaster, assisted by Mr. Edgar Walton, organist. “Break Forth into Joy ” (Caleb Simper), by the full choir, was a feature. In the singing of the hymns the congregation joined whole-heartedly. The party adjoined to the Foresters ’ Hall, where a sumptuous repast was provided. A social gathering and public meeting were held in the evening....At intervals in the addresses, vocal and instrumental items were given. On Sunday next Rev. George WaIters will preach in the new church. (Auckland Star, Fri. Dec. 6, 1901.) OPENING OF CHURCH. The 4th of December 1901, will remain a historic day for us. Such a magnificent beginning surpassed our wildest hopes. It was proof that we have not only a firm root in Auckland,...
Regression Diagnostics John Maindonald October 7, 2013 1 Calibration of Regression Diagnostics Indications of departures from regression assumptions in diagnostic plots may reflect sampling variation. This is an especial issue for... more
Regression Diagnostics John Maindonald October 7, 2013 1 Calibration of Regression Diagnostics Indications of departures from regression assumptions in diagnostic plots may reflect sampling variation. This is an especial issue for relatively small datasets. Diagnostic plots for a number of sets of simulated data may be an essential aid to judgement. In e↵ect, the observed diagnostic plot is judged against a simulated sampling distribution for such plots.
Never HIGHLIGHT a Book Again! Includes all testable terms, concepts, persons, places, and events. Cram101 Just the FACTS101 studyguides gives all of the outlines, highlights, and quizzes for your textbook with optional online... more
Never HIGHLIGHT a Book Again! Includes all testable terms, concepts, persons, places, and events. Cram101 Just the FACTS101 studyguides gives all of the outlines, highlights, and quizzes for your textbook with optional online comprehensive practice tests
Research Interests:
Languages shape the way we think, and determine what we can think about. [Benjamin Whorf.] S has forever altered the way people analyze, visualize, and manipulate data... S is an elegant, widely accepted, and enduring software system,... more
Languages shape the way we think, and determine what we can think about. [Benjamin Whorf.] S has forever altered the way people analyze, visualize, and manipulate data... S is an elegant, widely accepted, and enduring software system, with conceptual integrity, thanks to the insight, taste, and effort of John Chambers.
Research Interests:
Ordination is a generic name for methods for providing a low-dimensionaL view of points in multi-dimensional space, such that “similar” objects are near each other and dissimilar objects are separated. The plot(s) from an ordination in 2... more
Ordination is a generic name for methods for providing a low-dimensionaL view of points in multi-dimensional space, such that “similar” objects are near each other and dissimilar objects are separated. The plot(s) from an ordination in 2 or 3 dimensions may provide useful visual clues on clusters in the data and on outliers. If data can be separated into known classes that should be reflected in any ordination, then the scores from classification using lda() may be a good basis for an ordination. Plots in 2 or perhaps 3 dimensions may then reveal additional classes and/or identify points that may be misclassified and/or are in some sense outliers. It may indicate whether the classes that formed the basis for the ordination seem real and/or the effectiveness of the discrimination method in choosing the boundaries between classes. Here, the discussion will turn to multi-dimensional scaling (MDS) methods where distances are given, or that start by calculating distances between points, ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
... STEPHEN P. FOSTER,2 JAMIE VAN DEN HEUVEL,! ... pmticularly the introduced Australian species, Iightbrown apple moth, Epiphyas postvittmw (Walker), and the two endemic New Zealand species, Planotortrix octo Dugdale and Ctenopseustis... more
... STEPHEN P. FOSTER,2 JAMIE VAN DEN HEUVEL,! ... pmticularly the introduced Australian species, Iightbrown apple moth, Epiphyas postvittmw (Walker), and the two endemic New Zealand species, Planotortrix octo Dugdale and Ctenopseustis obliquana (Walker) (Waddell et al ...
Research Interests:
... DIANA C. WHITING, GLENDA M. O'CONNOR, JAMIE VANDEN HEUVEL, AND JOHN II. ... Payne, RW, PW Lane, AE Ainsley, KE Bick-neD, PCN Digby, SA Harding, PK Leech, H. R. Simpson, AD Todd, SJ Welham, RP Whitc, R. Thompson, CT Wilson, LG... more
... DIANA C. WHITING, GLENDA M. O'CONNOR, JAMIE VANDEN HEUVEL, AND JOHN II. ... Payne, RW, PW Lane, AE Ainsley, KE Bick-neD, PCN Digby, SA Harding, PK Leech, H. R. Simpson, AD Todd, SJ Welham, RP Whitc, R. Thompson, CT Wilson, LG Under-hill, and PJ Verrier ...
Research Interests:
J. Econ. Entomol. 88(5): 1356-1360 (1995) ABSTRACT Mortality responses to hot water at 39--47°Cwere determined for 1-, 4-, and 7- d-old eggs and 1st, 3rd, and 5th instars of 3 tortricid pests, the Australian species Epiphyas postvittana... more
J. Econ. Entomol. 88(5): 1356-1360 (1995) ABSTRACT Mortality responses to hot water at 39--47°Cwere determined for 1-, 4-, and 7- d-old eggs and 1st, 3rd, and 5th instars of 3 tortricid pests, the Australian species Epiphyas postvittana (Walker), and 2 species endemic to ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
ABSTRACT First-instar light brown apple moths, Epiphyas postvittana (Walker), were exposed to various controlled-atmosphere treatments at different temperatures. The time required for 99% mortality (LT••) was highly dependent on the... more
ABSTRACT First-instar light brown apple moths, Epiphyas postvittana (Walker), were exposed to various controlled-atmosphere treatments at different temperatures. The time required for 99% mortality (LT••) was highly dependent on the concentration of 0, and temperature; insects died more rapidly with lower concentrations of O, and with higher temperatures. The effect of different concentrations of CO, on mortality of first instars was less pronounced. At 2O"C and an 0, concentration of 0.4%, a CO, concentration of 5.0% resulted in lowest LT•• (mean of 22.5 h) of all of the CO, concentrations tested. Using this atmosphere (i.e., 0.4% 0" 5.0% CO,), we determined mortality of third instars, fifth instars, and eggs of E. postvtttana at 20 and 300c. At both temperatures, LTgos decreased in the following order: fifth instar > third instar "o" eggs > first instar. An experiment testing this atmosphere at 40"C on fifth instars gave a mean LT•• of 4.2 h.
Research Interests:
J. Econ. Entomol.85(4): 1222--1230 (1992) ABSTRACT Fumigation experiments with methyl bromide (MeBr) were done in New Zealand over the summers at the end of 1987 and 1989. A total of 74,033 freshly laid codling moth, Cydia pomonella (L.),... more
J. Econ. Entomol.85(4): 1222--1230 (1992) ABSTRACT Fumigation experiments with methyl bromide (MeBr) were done in New Zealand over the summers at the end of 1987 and 1989. A total of 74,033 freshly laid codling moth, Cydia pomonella (L.), eggs on the cherry ...
Research Interests:
Research Interests:
Research Interests:
... Apples DIANA C. WHITING,l GLENDA M. O'CONNOR,l ANDJOHN H. MAINDONALD2 Environ ... 169. Auck-land.NewZealand. 2Departmentof Statistics.Universityof Newcastle,Callaghan, NewSouthWales2308, Australia. tation ...
Research Interests:
... R. DENTENER,t SONYA E. LEWfHWAITE\ JOHN H. MAINDONALD,2 ANDPATRICK G. CONNOLLY] ... for assistancewith statisticalanaly-ses; and Ngaire Markwick,Howard Wearing, andMichael Lay-Yeefor ... Dentener, PR, K V. Bennett, LE Hoy, SE... more
... R. DENTENER,t SONYA E. LEWfHWAITE\ JOHN H. MAINDONALD,2 ANDPATRICK G. CONNOLLY] ... for assistancewith statisticalanaly-ses; and Ngaire Markwick,Howard Wearing, andMichael Lay-Yeefor ... Dentener, PR, K V. Bennett, LE Hoy, SE Lewthwaite, PJ Lestet·,JH ...
Research Interests:
The term algorithm is generally used to describe a series of logical steps that, when implemented within a computer program, will perform a desired computational task. The task must be defined precisely. Issues that are generally... more
The term algorithm is generally used to describe a series of logical steps that, when implemented within a computer program, will perform a desired computational task. The task must be defined precisely. Issues that are generally important in finding or choosing such an algorithm include: computational cost, computer memory or storage requirements, accuracy and precision, simplicity of description, and the tuning of the algorithm to computer architecture and organization. Keywords: algorithm; program; computational complexity; program logic; subroutine
Research Interests:
Computer systems that hold data of a private or confidential nature should be designed so that data are “disclosed only to authorized persons, entities and processes at authorized times and in the authorized manner” 24. Rights of privacy... more
Computer systems that hold data of a private or confidential nature should be designed so that data are “disclosed only to authorized persons, entities and processes at authorized times and in the authorized manner” 24. Rights of privacy and confidentiality may be enshrined in legislation. Such rights are typically not absolute; they may, especially where data is required for research purposes, be weighed against a wider public interest. Demands for privacy and confidentiality impose responsibilities on designers and managers of computer systems that store sensitive data. Advances in computer technology, and especially the use of the internet, have created huge security challenges. Keywords: confidentiality; privacy; data base
Research Interests:
Floating point number representations are a response to the demand to store and manipulate in a computer, using a fixed number of digits, numbers that range widely in magnitude. The consequent finite precision of floating point arithmetic... more
Floating point number representations are a response to the demand to store and manipulate in a computer, using a fixed number of digits, numbers that range widely in magnitude. The consequent finite precision of floating point arithmetic has large implications for practical computation, of a kind that this article explores. Keywords: computer arithmetic; accuracy; precision; floating point; fixed point; IEEE arithmetic
Research Interests:
Computer systems that hold data of a private or confidential nature should be designed so that data are “disclosed only to authorized persons, entities and processes at authorized times and in the authorized manner” 24. Rights of privacy... more
Computer systems that hold data of a private or confidential nature should be designed so that data are “disclosed only to authorized persons, entities and processes at authorized times and in the authorized manner” 24. Rights of privacy and confidentiality may be enshrined in legislation. Such rights are typically not absolute; they may, especially where data is required for research purposes, be weighed against a wider public interest. Demands for privacy and confidentiality impose responsibilities on designers and managers of computer systems that store sensitive data. Advances in computer technology, and especially the use of the internet, have created huge security challenges. Keywords: confidentiality; privacy; data base
Research Interests:
Research Interests:
Computer languages are artificial languages that enable humans to give instructions to computer systems. Texts written in these languages are called computer programs or code. Influences that explain differences between languages include... more
Computer languages are artificial languages that enable humans to give instructions to computer systems. Texts written in these languages are called computer programs or code. Influences that explain differences between languages include differences of historical development, of conceptual framework, and of intended use. The use of computer languages for writing large and complex programs raises major issues of program design. The use of computer languages for statistical analysis raises further issues that have been the subject of extensive discussion and research. Keywords: language; program; compiler; interpreter
Research Interests:
We describe a normalization procedure for the score of a pairwise alignment of two biological sequences and an algorithm (POZITIV) that nds the normalized score ecien tly. Designed to overcome the bias due to the composition of the... more
We describe a normalization procedure for the score of a pairwise alignment of two biological sequences and an algorithm (POZITIV) that nds the normalized score ecien tly. Designed to overcome the bias due to the composition of the alignment, the normalized score (which we call the POZ score) measures the distance (in standard deviations) between Smith-Waterman (S-W) score of the aligned letters and the mean value of all other S-W scores obtained by a permutation of either sequence. Unlike the usual Z-score calculation used in sequence alignment, the mean and standard deviation are taken without realigning the permuted sequences. We demonstrate that calculating the mean and standard deviation in this way enables us to calculate exactly and in two steps, the rst being O(N) time, where N is the length of the sequence, the second in a xed number of calculations i.e. in O(1) time.
Research Interests:
Research Interests:
This document identifies statistical issues that can be and commonly are important for data mining problems. As far as possible, it will avoid the technical language of mathematical statistics. Key issues for any data analysis are: 1. Why... more
This document identifies statistical issues that can be and commonly are important for data mining problems. As far as possible, it will avoid the technical language of mathematical statistics. Key issues for any data analysis are: 1. Why are we undertaking this investigation? 2. What is the intended use of results? 3. What limitations, arising from the manner of collection or from the incompleteness of the information, may constrain that intended use? Finally, when results are presented, the data analyst should be well placed to answer the question: "What is the relevance of these results?" Part I discusses statistical issues and ideas. Examples, most of them taken from published data, highlight the importance of these issues. Remaining parts of the document summarize important results and concepts from classical statistics. More recent data mining methodologies supplement rather than displace the classical methodologies. Models used in data mining often build on or incop...
Research Interests:
This module will present a statistical perspective on data mining problems. It will, of necessity, do this without the requirement for a strong background in statistical ideas. It will be clear that I have been strongly influenced by my... more
This module will present a statistical perspective on data mining problems. It will, of necessity, do this without the requirement for a strong background in statistical ideas. It will be clear that I have been strongly influenced by my reading of the data mining literature, while at the same time taking issue with some claims that are common in that literature. Components of this module will be: The R System: An introduction to the R system, for data analysis, graphics and scientific computing generally. This gives access to a huge range of abilities that can be used for data manipulation and data analysis. The aim will be to give a limited facility with some of the more commonly used tools, both in the statistical community and in the data mining community. What is the purpose of these analyses?: An overview of answers that may typically be given to the question: "Why are we undertaking this investigation?", and of implications that those answers have for the data analys...
Research Interests:
Research Interests:
of Contents ii If we teach only the findings and products of science – no matter how useful and even inspiring they may be – without communicating its critical method, how can the average person possibly distinguish science from... more
of Contents ii If we teach only the findings and products of science – no matter how useful and even inspiring they may be – without communicating its critical method, how can the average person possibly distinguish science from pseudoscience? . . . Many, perhaps most, textbooks for budding scientists tread lightly here. It is enormously easier to present in an appealing way the wisdom distilled from centuries of patient and collective interrogation of nature than to detail the messy distillation apparatus. The method of science, as stodgy and grumpy as it may seem, is far more important than the findings of science.
Research Interests:
Primarily, the purpose of this document is to note issues for the use of linear and gen-eralized linear models, and of other regression models. Worked examples, as far as possible using data that have been a basis for published research,... more
Primarily, the purpose of this document is to note issues for the use of linear and gen-eralized linear models, and of other regression models. Worked examples, as far as possible using data that have been a basis for published research, are used as a basis for discussion of the following issues: Missing variables; noting very striking examples that arise in multi-way tables, perhaps modeled using logistic regression; [Maindonald & Braun (2007, Subsections 2.2.1, 3.4.5, 6.8.3 & Section 8.3)] Observational versus experimental data – implications for interpretation and inference; Maindonald & Braun (2007, Chapter 6); Rosenbaum (2002)] Variable selection, noting the use of resampling methods to obtain realistic "error" estimates; [Maindonald & Braun (2007, Chapter 6)] Errors in explanatory variables; implications of classical measurement error for infer-ence; [Maindonald & Braun (2007, Chapter 6); Carroll (2006, Chapter 1)] Regression on constructed variables – propensity sco...
These notes review the theory of linear models, demonstrating the utility of Householder reflections, the QR decomposition and the Cholesky decomposition both for computation and for theory; [c.f., Chapter 1 of Wood (2006)] introduce... more
These notes review the theory of linear models, demonstrating the utility of Householder reflections, the QR decomposition and the Cholesky decomposition both for computation and for theory; [c.f., Chapter 1 of Wood (2006)] introduce computational, theoretical and practical aspects of regression splines, smooth-ing splines and penalized splines. This leads into an introduction to Generalized Ad-ditive models, limiting attention however to smooth functions of a single variable. [Maindonald & Braun (2010, Section 7.5, pp. 231–238)] demonstrate the use of resmampling and simulation methods for comparing models; [See Subsection 5.3 for references] give brief summary details of the theory of Generalized Linear Models, with logistic regression and Poisson regression as special cases. The extension into Generalized Additive models with an exponential error distribution follows naturally.
Research Interests:
Computer languages are artificial languages that enable humans to give instructions to computer systems. Texts written in these languages are called computer programs or code. Influences that explain differences between languages include... more
Computer languages are artificial languages that enable humans to give instructions to computer systems. Texts written in these languages are called computer programs or code. Influences that explain differences between languages include differences of historical development, of conceptual framework, and of intended use. The use of computer languages for writing large and complex programs raises major issues of program design. The use of computer languages for statistical analysis raises further issues that have been the subject of extensive discussion and research. Keywords: language; program; compiler; interpreter
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests: Biology and Microarray
Research Interests:
Research Interests:
Research Interests:
... Diana C. Whiting Corresponding Author Contact Information , Glenda M. O'Connor andJohn H. Maindonald. ... J. De Baerdemaeker,B. Mc Kenna, M. Janssens, A. Thompson, R Artes Calero, E. Hhn and Z. Somogyi(Eds.), 1995.Quality... more
... Diana C. Whiting Corresponding Author Contact Information , Glenda M. O'Connor andJohn H. Maindonald. ... J. De Baerdemaeker,B. Mc Kenna, M. Janssens, A. Thompson, R Artes Calero, E. Hhn and Z. Somogyi(Eds.), 1995.Quality Criteria. ...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
The Lee-Carter method of mortality forecasting assumes an invariant age component and most applications have adopted a linear time component. The use of the method with Australian data is compromised by significant departures from... more
The Lee-Carter method of mortality forecasting assumes an invariant age component and most applications have adopted a linear time component. The use of the method with Australian data is compromised by significant departures from linearity in the time component and changes over time in the age component. We modify the method to adjust the time component to reproduce the age distribution of deaths, rather than total deaths, and to determine the optimal fitting period in order to address non-linearity in the time component. In the Australian case the modification has the added advantage that the assumption of invariance is better met. For Australian data, the modifications result in higher forecast life expectancy than the original Lee-Carter method and official projections, and a 50 per cent reduction in forecast error. The model is also expanded to take account of age-time interactions by incorporating additional terms, but these are not readily incorporated into forecasts.
Research Interests:
ABSTRACT
Research Interests:
ABSTRACT Presentation issues provide a perspective from which to comment on the role of statistics in applied biological research. Full details are required of the experimental layout, and of reasons for the choice of design. Gains from... more
ABSTRACT Presentation issues provide a perspective from which to comment on the role of statistics in applied biological research. Full details are required of the experimental layout, and of reasons for the choice of design. Gains from effective design may include: improved precision and clearcut results; reduced cost and/or effort; information that a less satisfactory design could not have provided; and clues that will help in planning further experiments. Failure to distinguish distinct strata of variation is a common source of error, both for design and for analysis. Valid treatment comparisons require replication at the level of whole treatment units. Points that require attention in analysing data and reporting results include: limitations in the design and in the data; evidence of appropriate checks; the tying of statistical evidence to biologically meaningful questions; translation of results back into subject matter terms; and lucid presentation. Analyses must reflect the structure of the data; thus a trend or response surface analysis is appropriate when levels of one or more factors vary quantitatively. Attention is drawn to some major recent advances in approaches to design and analysis.
Research Interests:
Research Interests:
Research Interests: Computer Science, Statistics, Agronomy, Statistical Computing, Data Analysis, and 15 moreStatistical Analysis, Plant Biology, Biometrics, Experimental Design, Structured data, Checklist, Data Transformation, Design of experiment, Soil sciences, Experimentation, Application, Scientific Research, Standard Error, Confidence Interval, and SIgnificance test
ABSTRACT Various models that have been proposed to describe the avoidance of super-parasitism in single ovipositing parasite-host systems are surveyed and presented within a framework that clarifies the assumptions. Computations on the... more
ABSTRACT Various models that have been proposed to describe the avoidance of super-parasitism in single ovipositing parasite-host systems are surveyed and presented within a framework that clarifies the assumptions. Computations on the information statistics are summarized, and a new procedure, which is a variant of a sequential occupancy model, is illustrated on data for wasps laying eggs in some housefly pupae; there are also comparative analyses with some older methods. All models can be formulated within the context of the time-dependent behaviour of a Markov process. Within this context inferences concerning the parameter(s) measuring any avoidance of superparasitism can readily be made conditional on the observed number of ovipositions, thus avoiding estimation of the expected number of ovipositions as a nuisance parameter.
Research Interests:
Research Interests: Geography, Epidemiology, Climate Change, Climate variability, Forecasting, and 15 moreMedicine, Disease Outbreaks, global Climate change, Humans, Dengue fever, Animals, Dengue, Lancet, Human health, Greenhouse Effect, Humidity, Logistic Models, Empirical Model, Aedes, and Medical and Health Sciences
The following topics will be treated at various levels of detail, depend- ing on time available. Review the theory of linear models, noting the utility of Householder reections,and the QR decomposition both for computation and for theory.... more
The following topics will be treated at various levels of detail, depend- ing on time available. Review the theory of linear models, noting the utility of Householder reections,and the QR decomposition both for computation and for theory. [c.f., Chapter 1 of Wood (2006)] Discuss regression splines, noting in passing the further extension into generalized additive models. Maindonald & Braun (2007, Section 7.1, pp. 234{238). Introduce the theory of Generalized Linear Models, with logistic re- gression and Poisson regression as special cases. [Maindonald & Braun (2007, Sections 8.1 { 8.5); Wood (2006, Sec- tions 2.1 { 2.3)] Brief discussion of supervised learning (classication), using tree- based methods, Breiman’s random forests, and linear discriminant analysis as implemented,in R’s function lda(). The interest here in the use of these methods,to provide constructed variables (in this context, known as \propensities") for use as explanatory variables in regression. [Maindonald &...
Research Interests:
The temperature sensitivity of three egg ages of Plodia interpunctella (Hübner) were examined at 42–48°C heat (hot water) only treatments, 0.5–11.5°C cold storage only treatments, or combination treatments of 42 or 46°C, followed by... more
The temperature sensitivity of three egg ages of Plodia interpunctella (Hübner) were examined at 42–48°C heat (hot water) only treatments, 0.5–11.5°C cold storage only treatments, or combination treatments of 42 or 46°C, followed by subsequent 0–10.5°C cold storage. The 1-day-old egg was typically more tolerant of heat only treatments than the 2- and 3-day-old egg. A 99% mortality of all
Research Interests:
Research Interests:
Research Interests:
1 The architecture and leaf dimensions (petiole length, blade length and blade width) of 403 tree species were studied at six sites ranging from lowland equatorial to temperate Australian rain forest. The branching patterns of understorey... more
1 The architecture and leaf dimensions (petiole length, blade length and blade width) of 403 tree species were studied at six sites ranging from lowland equatorial to temperate Australian rain forest. The branching patterns of understorey saplings and exposed canopy trees ...
Research Interests:
Research Interests:
Research Interests:
Research Interests: Biomedical Engineering, Biology, Medicine, Linear models, Spleen, and 15 moreAnatomy, Kidney, Liver, Allometry, Animals, Male, Heart, Medical Physiology, Fur Seal, Lung, Intestines, Fur Seals, Body Weight, Stomach, and Viscera
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
To ensure that the best scientific evidence is available to guide conservation action, effective mechanisms for communicating the results of research are necessary. In medicine, an evidence-based approach assists doctors in applying... more
To ensure that the best scientific evidence is available to guide conservation action, effective mechanisms for communicating the results of research are necessary. In medicine, an evidence-based approach assists doctors in applying scientific evidence when treating patients. The approach has required the development of new methods for systematically reviewing research, and has led to the establishment of independent organizations to disseminate the conclusions of reviews. (1) Such methods could help bridge gaps between researchers and practitioners of environmental conservation. In medicine, systematic reviews place strong emphasis on reviewing experimental clinical trials that meet strict standards. Although experimental studies are much less common in conservation, many of the components of systematic reviews that reduce the biases when identifying, selecting and appraising relevant studies could still be applied effectively. Other methods already applied in medicine for the revi...
Research Interests: Management, Science Communication, Clinical Trial, Environmental Management, Biology, and 15 moreCausation, Ecology, Landscape, Biological Sciences, Environmental Conservation, Environmental Sciences, Method, Sustainable Use, Experimental Study, Systematic review, Populations, Implementation, Dissemination, Adaptive Management, and Environmental
Background Anthozoan cnidarians are amongst the simplest animals at the tissue level of organization, but are surprisingly complex and vertebrate-like in terms of gene repertoire. As major components of tropical reef ecosystems, the stony... more
Background Anthozoan cnidarians are amongst the simplest animals at the tissue level of organization, but are surprisingly complex and vertebrate-like in terms of gene repertoire. As major components of tropical reef ecosystems, the stony corals are anthozoans of particular ecological significance. To better understand the molecular bases of both cnidarian development in general and coral-specific processes such as skeletogenesis and symbiont acquisition, microarray analysis was carried out through the period of early development – when skeletogenesis is initiated, and symbionts are first acquired. Results Of 5081 unique peptide coding genes, 1084 were differentially expressed (P ≤ 0.05) in comparisons between four different stages of coral development, spanning key developmental transitions. Genes of likely relevance to the processes of settlement, metamorphosis, calcification and interaction with symbionts were characterised further and their spatial expression patterns investigat...
Research Interests: Evolutionary Biology, Biology, Medicine, Gene expression, Biological Sciences, and 15 moreAnthozoa, Expressed Sequence Tags, Animals, Cluster Analysis, Coral Reef, Early development, Gene, Coral, Microarray Analysis, Large Scale, Amino Acid Sequence, Gene expression profiling, Differential expression, Medical and Health Sciences, and Expression pattern
This article discusses the results in Boys and Henderson (2004, Biometrics 60, 573-581) in which the authors propose a new approach to the classification of genomic DNA into a number of hidden Markov states with a variable order of... more
This article discusses the results in Boys and Henderson (2004, Biometrics 60, 573-581) in which the authors propose a new approach to the classification of genomic DNA into a number of hidden Markov states with a variable order of dependency, potentially allowing for the high-throughput detection of structure within genomic DNA. This article is likely to be an important point of departure for further modeling of this type. We question whether the genome of the bacteriophage lambda is the most appropriate example with which to demonstrate the method's effectiveness, whether it can be expected that the method will carry over to genomes where there is only one direction of transcription and no operon structure, and suggest a graphical display that seems to offer insight into the results. It would be interesting to see an analysis that uses the codon alphabet.
Research Interests:
Research Interests:
Research Interests:
Page 1. Data Analysis and Graphics Using R Draft of Chapter 1 for the Second Edition JH Maindonald and WJ Braun c JH Maindonald and WJ Braun 2005. Permission is given to make copies for personal study. Any other use is strictly... more
Page 1. Data Analysis and Graphics Using R Draft of Chapter 1 for the Second Edition JH Maindonald and WJ Braun c JH Maindonald and WJ Braun 2005. Permission is given to make copies for personal study. Any other use is strictly prohibited. 1 Page 2. 1 ...
Research Interests:
Application of the Lee-Carter method to Australian data shows that model assumptions are not always met because of age-time interactions. The Lee-Carter method is adapted to take account of departures from linearity in the dominant time... more
Application of the Lee-Carter method to Australian data shows that model assumptions are not always met because of age-time interactions. The Lee-Carter method is adapted to take account of departures from linearity in the dominant time component and the failure to satisfy the ...