Note: Supplemental materials are not guaranteed with Rental or Used book purchases.
Purchase Benefits
What is included with this book?
E. Marian Scott, Professor of Environmental Statistics, Department of Statistics, University of Glasgow
Marian Scott is the editor of the Earth & Environmental Sciences section of Statistics in Practice for Wiley. She has 49 published articles. Her reputation within the community is evidenced by her associate editorship for 2 external journals (Radiocarbon and the Journal of Environmental Radioactivity); she is also honorary secretary for the Royal Statistical Society's Environmental Statistics Study Group.
Preface | p. xi |
Contributing authors | p. xiii |
Methodology | p. 1 |
Introduction | p. 3 |
What is a trend? | p. 4 |
Why analyse trends? | p. 5 |
Some simple examples | p. 6 |
Dutch wind speeds | p. 6 |
North Sea haddock stocks | p. 8 |
Alkalinity in the Round Loch of Glenhead | p. 10 |
Atmospheric ozone in eastern England | p. 12 |
Considerations and difficulties | p. 14 |
Autocorrelation | p. 14 |
Effect of other variables | p. 15 |
Lack of designed experiments | p. 15 |
Consideration of auxiliary information | p. 16 |
The necessity of extrapolation | p. 17 |
Scope of the book | p. 19 |
Further reading | p. 20 |
References | p. 21 |
Exploratory analysis | p. 25 |
Data visualisation | p. 25 |
Time series plots | p. 26 |
Boxplots | p. 28 |
The autocorrelation function | p. 30 |
Irregularly spaced data - the variogram | p. 35 |
Relationships between variables | p. 38 |
Simple smoothing | p. 41 |
Moving averages | p. 41 |
Local polynomial fitting | p. 42 |
Further considerations | p. 43 |
Linear filters | p. 45 |
Frequency considerations | p. 46 |
The convolution theorem and filter design | p. 48 |
Dealing with end effects | p. 50 |
Other applications | p. 51 |
Classical test procedures | p. 54 |
Concluding comments | p. 57 |
References | p. 58 |
Parametric modelling - deterministic trends | p. 61 |
The linear trend | p. 63 |
Checking the model assumptions | p. 65 |
Choosing the time index | p. 69 |
An overview of least squares | p. 70 |
Extrapolation | p. 74 |
Influential observations | p. 76 |
Other methods of model fitting | p. 79 |
Multiple regression techniques | p. 82 |
Representing seasonality in regression models | p. 83 |
Interactions | p. 86 |
Model building and selection | p. 87 |
Violations of assumptions | p. 94 |
Dealing with heteroscedasticity | p. 94 |
Dealing with non-normality | p. 96 |
Dealing with autocorrelation | p. 97 |
Nonlinear trends | p. 105 |
Nonlinear least squares | p. 105 |
Cycles | p. 106 |
Changepoints and interventions | p. 108 |
Generalised linear models | p. 111 |
Parameter estimation | p. 113 |
Model comparison | p. 115 |
Model checking | p. 116 |
Prediction with GLMs | p. 117 |
Extensions and refinements | p. 118 |
Inference with small samples | p. 120 |
References | p. 122 |
Nonparametric trend estimation | p. 127 |
An introduction to nonparametric regression | p. 127 |
Linear smoothing | p. 128 |
Local linear regression | p. 129 |
Spline smoothing | p. 130 |
Choice of smoothing parameter | p. 131 |
Variance estimators | p. 135 |
Standard errors for the regression function | p. 135 |
Testing for consistency with parametric models | p. 136 |
Multiple covariates | p. 140 |
Additive, semiparametric and bivariate models | p. 140 |
The backfitting algorithm | p. 142 |
Inference for additive models | p. 144 |
Handling autocorrelation | p. 148 |
Other nonparametric estimation techniques | p. 151 |
Lowess smoothing | p. 151 |
Wavelets | p. 154 |
Varying coefficient models | p. 160 |
Discontinuity detection | p. 161 |
Quantile regression | p. 162 |
Parametric or nonparametric? | p. 166 |
References | p. 167 |
Stochastic trends | p. 171 |
Stationary time series models and their properties | p. 171 |
Autoregressive processes | p. 171 |
Moving average processes | p. 174 |
Mixed ARMA processes | p. 174 |
Model identification | p. 175 |
Parameter estimation | p. 177 |
Model checking | p. 182 |
Forecasting | p. 186 |
The backshift operator | p. 190 |
Trend removal via differencing | p. 193 |
ARIMA models | p. 194 |
Spurious regressions | p. 199 |
Long memory models | p. 201 |
Models for irregularly spaced series | p. 205 |
State space and structural models | p. 207 |
Simple structural time series models | p. 207 |
The state space representation | p. 209 |
The Kalman filter | p. 213 |
Parameter estimation | p. 217 |
Connection with nonparametric smoothing | p. 219 |
Nonlinear models | p. 228 |
References | p. 231 |
Other issues | p. 235 |
Multisite data | p. 235 |
Visualisation | p. 236 |
Modelling | p. 239 |
Multivariate series | p. 241 |
Dimension reduction | p. 241 |
Multivariate models | p. 244 |
Point process data | p. 245 |
Poisson processes | p. 246 |
Other point process models | p. 249 |
Marked point processes | p. 250 |
Trends in extremes | p. 250 |
Approaches based on block maxima | p. 251 |
Approaches based on threshold exceedances | p. 253 |
Modern developments | p. 256 |
Censored data | p. 257 |
References | p. 260 |
Case Studies | p. 265 |
Additive models for sulphur dioxide pollution in Europe | p. 267 |
Introduction | p. 267 |
Additive models with correlated errors | p. 269 |
An introduction to additive models | p. 269 |
Smoothing techniques | p. 270 |
Smoothing correlated data | p. 272 |
Fitting additive models | p. 273 |
Comparing nonparametric models | p. 276 |
Models for the SO2 data | p. 277 |
Conclusions | p. 281 |
Acknowledgement | p. 282 |
References | p. 282 |
Rainfall trends in southwest Western Australia | p. 283 |
Motivation | p. 283 |
The study region | p. 285 |
Data used in the study | p. 285 |
Modelling methodology | p. 289 |
Generalised linear models for daily rainfall | p. 289 |
Temporal and spatial dependence | p. 290 |
Covariates considered | p. 291 |
Modelling strategy | p. 292 |
Results | p. 293 |
Diagnostics | p. 294 |
Trends in wet-day precipitation amounts | p. 295 |
Trends in precipitation occurrence | p. 296 |
Combined trends in occurrence and amounts | p. 302 |
Summary and conclusions | p. 303 |
References | p. 304 |
Estimation of common trends for trophic index series | p. 307 |
Introduction | p. 307 |
Data exploration | p. 311 |
Common trends and additive modelling | p. 314 |
Adding autocorrelation to the additive model | p. 316 |
Combining the data from the eight stations | p. 319 |
Doing it all within a parametric model | p. 323 |
Dynamic factor analysis to estimate common trends | p. 324 |
The underlying model | p. 324 |
Discussion | p. 328 |
Acknowledgement | p. 329 |
References | p. 329 |
A space-time study on forest health | p. 333 |
Forest health: survey and data | p. 333 |
Regression models for longitudinal data with ordinal responses | p. 336 |
Spatiotemporal models | p. 340 |
Penalised splines | p. 341 |
Interaction surfaces | p. 343 |
Spatial trends | p. 344 |
Inference in spatiotemporal models | p. 346 |
Spatiotemporal modelling and analysis of forest health data | p. 348 |
Acknowledgements | p. 357 |
References | p. 357 |
Index | p. 359 |
Table of Contents provided by Ingram. All Rights Reserved. |
The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.
The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.