ALBERT

Note: Contents Contributors 1 Introduction 1.1 Part 1: Applied statistical theory 1.2 Part 2: The case studies 1.3 Data, software and flowcharts 2 Data management and software 2.1 Introduction 2.2 Data management 2.3 Data preparation 2.4 Statistical software 3 Advice for teachers 3.1 Introduction 4 Exploration 4.1 The first steps 4.2 Outliers, transformations and standardisations 4.3 A final thought on data exploration 5 Linear regression 5.1 Bivariate linear regression 5.2 Multiple linear regression 5.3 Partial linear regression 6 Generalised linear modelling 6.1 Poisson regression 6.2 Logistic regression 7 Additive and generalised additive modelling 7.1 Introduction 7.2 The additive model 7.3 Example of an additive model 7.4 Estimate the smoother and amount of smoothing 7.5 Additive models with multiple explanatory variables 7.6 Choosing the amount of smoothing 7.7 Model selection and validation 7.8 Generalised additive modelling 7.9 Where to go from here 8 Introduction to mixed modelling 8.1 Introduction 8.2 The random intercept and slope model 8.3 Model selection and validation 8.4 A bit of theory 8.5 Another mixed modelling example 8.6 Additive mixed modelling 9 Univariate tree models 9.1 Introduction 9.2 Pruning the tree 9.3 Classification trees 9.4 A detailed example: Ditch data 10 Measures of association 10.1 Introduction 10.2 Association between sites: Q analysis 10.3 Association among species: R analysis 10.4 Q and R analysis: Concluding remarks 10.5 Hypothesis testing with measures of association 11 Ordination — First encounter 11.1 Bray-Curtis ordination 12 Principal component analysis and redundancy analysis 12.1 The underlying principle of PCA 12.2 PCA: Two easy explanations 12.3 PCA: Two technical explanations 12.4 Example of PCA 12.5 The biplot 12.6 General remarks 12.7 Chord and Hellinger transformations 12.8 Explanatory variables 12.9 Redundancy analysis 12.10 Partial RDA and variance partitioning 12.11 PCA regression to deal with collinearity 13 Correspondence analysis and canonical correspondence analysis 13.1 Gaussian regression and extensions 13.2 Three rationales for correspondence analysis 13.3 From RGR to CCA13.4 Understanding the CCA triplot 13.5 When to use PCA, CA, RDA or CCA 13.6 Problems with CA and CCA 14 Introduction to discriminant analysis 14.1 Introduction 14.2 Assumptions 14.3 Example 14.4 The mathematics 14.5 The numerical output for the sparrow data 15 Principal coordinate analysis and non-metric multidimensional scaling 15.1 Principal coordinate analysis 15.2 Non-metric multidimensional scaling 16 Time series analysis — Introduction 16.1 Using what we have already seen before 16.2 Auto-regressive integrated moving average models with exogenous variables 17 Common trends and sudden changes 17.1 Repeated LOESS smoothing 17.2 Identifying the seasonal component 17.3 Common trends: MAFA 17.4 Common trends: Dynamic factor analysis 17.5 Sudden changes: Chronological clustering 18 Analysis and modelling of lattice data 18.1 Lattice data 18.2 Numerical representation of the lattice structure 18.3 Spatial correlation 18.4 Modelling lattice data 18.5 More exotic models 18.6 Summary 19 Spatially continuous data analysis and modelling 19.1 Spatially continuous data 19.2 Geostatistical functions and assumptions 19.3 Exploratory variography analysis 19.4 Geostatistical modelling: Kriging 19.5 A full spatial analysis of the bird radar data 20 Univariate methods to analyse abundance of decapod larvae 20.1 Introduction 20.2 The data 20.3 Data exploration 20.4 Linear regression results 20.5 Additive modelling results 20.6 How many samples to take? 20.7 Discussion 21 Analysing presence and absence data for flatfish distribution in the Tagus estuary, Portugal 21.1 Introduction 21.2 Data and materials 21.3 Data exploration 21.4 Classification trees 21.5 Generalised additive modelling 21.6 Generalised linear modelling 21.7 Discussion 22 Crop pollination by honeybees in Argentina using additive mixed modelling 22.1 Introduction 22.2 Experimental setup 22.3 Abstracting the information 22.4 First steps of the analyses: Data exploration 22.5 Additive mixed modelling 22.6 Discussion and conclusions 23 Investigating the effects of rice farming on aquatic birds with mixed modelling 23.1 Introduction 23.2 The data 23.3 Getting familiar with the data: Exploration 23.4 Building a mixed model 23.5 The optimal model in terms of random components 23.6 Validating the optimal linear mixed model 23.7 More numerical output for the optimal model 23.8 Discussion 24 Classification trees and radar detection of birds for North Sea wind farms 24.1 Introduction 24.2 From radars to data 24.3 Classification trees 24.4 A tree for the birds 24.5 A tree for birds, clutter and more clutter 24.6 Discussion and conclusions 25 Fish stock identification through neural network analysis of parasite fauna 25.1 Introduction 25.2 Horse mackerel in the northeast Atlantic 25.3 Neural networks 25.4 Collection of data 25.5 Data exploration 25.6 Neural network results 25.7 Discussion 26 Monitoring for change: Using generalised least squares, non-metric multidimensional scaling, and the Mantel test on western Montana grasslands 26.1 Introduction 26.2 The data 26.3 Data exploration 26.4 Linear regression results 26.5 Generalised least squares results 26.6 Multivariate analysis results 26.7 Discussion 27 Univariate and multivariate analysis applied on a Dutch sandy beach community 27.1 Introduction 27.2 The variables 27.3 Analysing the data using univariate methods 27.4 Analysing the data using multivariate methods 27.5 Discussion and conclusions 28 Multivariate analyses of South-American zoobenthic species — spoilt for choice 28.1 Introduction and the underlying questions 28.2 Study site and sample collection 28.3 Data exploration 28.4 The Mantel test approach 28.5 The transformation plus RDA approach 28.6 Discussion and conclusions 29 Principal component analysis applied to harbour porpoise fatty acid data 29.1 Introduction 29.2 The data 29.3 Principal component analysis 29.4 Data exploration 29.5 Principal component analysis results 29.6 Simpler alternatives to PCA 29.7 Discussion 30 Multivariate analyses of morphometric turtle data — size and shape 30.1 Introduction 30.2 The turtle data 30.3 Data exploration 30.4 Overview of classic approaches related to PCA 30.5 Applying PCA to the original turtle data 30.6 Classic morphometric data analysis approaches 30.7 A geometric morphometric approach 31 Redundancy analysis and additive modelling applied on savanna tree data 31.1 Introduction 31.2 Study area 31.3 Methods 31.4 Results 31.5 Discussion 32 Canonical correspondence analysis of lowland pasture vegetation in the humid tropics of Mexico 32.1 Introduction 32.2 The study area 32.3 The data 32.4 Data exploration 32.5 Canonical correspondence analysis results 32.6 African star grass 32.7 Discussion and conclusion 33 Estimating common trends in Portuguese fisheries landings 33.1 Introduction 33.2 The time series data 33.3 MAFA and DFA 33.4 MAFA results 33.5 DFA results 33.6 Discussion 34 Common trends in demersal communities on the Newfoundland-Labrador Shelf 34.1 Introduction 34.2 Data 34.3 Time series analysis 34.4 Discussion 35 Sea level change and salt marshes in the Wadden Sea: A time series analysis 35.1 Interaction between hydrodynamical and biological factors 35.2 The data 35.3 Data exploration 35.4 Additive mixed modelling 35.5 Additive mixed modelling results 35.6 Discussion 36 Time series analysis of Hawaiian waterbirds 36.1 Introduction 36.2 Endangered Hawaiian waterbirds 36.3 Data exploration 36.4 Three ways to estimate trends 36.5 Additive mixed modelling 36.6 Sudden breakpoints 36.7 Discussion 37 Spatial modelling of forest community features in the Volzhsko-Kamsky reserve 37.1 Introduction 37.2 Study area 37.3 Data exploration 37.4 Models of boreality without spatial auto-correlation 37.5 Models of boreality with spatial auto-correlation 37.6 Conclusion References Index