Atjaunināt sīkdatņu piekrišanu

Sampling Theory: For the Ecological and Natural Resource Sciences [Mīkstie vāki]

(Fisheries Investigation Chief, NOAA Southwest Fisheries Science Center, USA), (Reader, School of Mathematics, University of Edinburgh, UK), (Professor Emeritus, Humboldt State University, USA)
  • Formāts: Paperback / softback, 360 pages, height x width x depth: 245x188x18 mm, weight: 782 g
  • Izdošanas datums: 01-Oct-2019
  • Izdevniecība: Oxford University Press
  • ISBN-10: 0198815808
  • ISBN-13: 9780198815808
Citas grāmatas par šo tēmu:
  • Mīkstie vāki
  • Cena: 79,42 €
  • Grāmatu piegādes laiks ir 3-4 nedēļas, ja grāmata ir uz vietas izdevniecības noliktavā. Ja izdevējam nepieciešams publicēt jaunu tirāžu, grāmatas piegāde var aizkavēties.
  • Daudzums:
  • Ielikt grozā
  • Piegādes laiks - 4-6 nedēļas
  • Pievienot vēlmju sarakstam
Sampling Theory: For the Ecological and Natural Resource SciencesPalielināt
  • Formāts: Paperback / softback, 360 pages, height x width x depth: 245x188x18 mm, weight: 782 g
  • Izdošanas datums: 01-Oct-2019
  • Izdevniecība: Oxford University Press
  • ISBN-10: 0198815808
  • ISBN-13: 9780198815808
Citas grāmatas par šo tēmu:
Permanent link: https://www.kriso.lv/db/9780198815808.html
Keywords:
This introductory text is specifically targeted at ecologists and resource scientists, illustrating how sampling theory can be applied in a wide variety of resource contexts. The emphasis throughout is on design-based sampling from finite populations but attention is also given to model-based prediction and sampling from infinite populations.

Sampling theory considers how methods for selection of a subset of units from a finite population (a sample) affect the accuracy of estimates of descriptive population parameters (mean, total, proportion). Although a sound knowledge of sampling theory principles would seem essential for ecologists and natural resource scientists, the subject tends to be somewhat overlooked in contrast to other core statistical topics such as regression analysis, experimental design, and multivariate statistics. This introductory text aims to redress this imbalance by specifically targeting ecologists and resource scientists, and illustrating how sampling theory can be applied in a wide variety of resource contexts. The emphasis throughout is on design-based sampling from finite populations, but some attention is given to model-based prediction and sampling from infinite populations.

Recenzijas

I echo what the authors identified as a general lack of formal training into sampling theory. Sampling Theory has filled a gap in my lack of knowledge surrounding certain designs and analytical concepts, so this book should also be a good resource for applied ecologists and researchers, and as the primary text in a graduate-level course on sampling design. * Stephen L. Webb, Journal of Wildlife Management * A good introduction to a difficult subject . . . The volume is easy to read if one can handle the mathematics and statistical theory required. There are many interesting examples (I especially like the one about hunting mushrooms) to motivate readers. Many exercises are also given at the end of most chapters. * F. James Rohlf, Stony Brook University, The Quarterly Review of Biology *

1 Introduction 1(10)
1.1 The design-based paradigm
2(2)
1.2 Text content and orientation
4(1)
1.3 What distinguishes this text?
5(1)
1.4 Recommendations for instructors
6(1)
1.5 Sampling theory: A brief history
7(4)
2 Basic concepts 11(12)
2.1 Terminology
12(1)
2.2 Components of a sampling strategy
13(1)
2.3 Selection methods
14(1)
2.4 Properties of estimators
15(3)
2.5 Sampling distribution of an estimator
18(1)
2.6 Judgment sampling versus random sampling
19(4)
3 Equal probability sampling 23(25)
3.1 Without replacement sampling
23(9)
3.1.1 Estimation of the population mean, proportion, and total
25(2)
3.1.2 Sampling variance
27(3)
3.1.3 Estimation of sampling variance
30(1)
3.1.4 Bernoulli sampling
31(1)
3.2 With replacement sampling
32(4)
3.2.1 Estimation of the population mean, proportion, and total
34(1)
3.2.2 Sampling variance and variance estimation
35(1)
3.2.3 Rao-Blackwell theorem
35(1)
3.3 Relative performance of alternative sampling strategies
36(2)
3.3.1 Measures of relative performance
36(1)
3.3.2 An example: SRS/mean-per-unit estimation versus SWR
37(1)
3.4 Sample size to achieve desired level of precision
38(4)
3.4.1 Approximate normality of sampling distributions
38(2)
3.4.2 Confidence interval construction
40(1)
3.4.3 Sample size determination
40(2)
3.5 Nonresponse and oversampling
42(2)
3.6 Sampling in R
44(1)
3.6.1 SRS and SWR
44(1)
3.6.2 Sample spaces
44(1)
3.7
Chapter comments
45(1)
Problems
45(3)
4 Systematic sampling 48(20)
4.1 Linear systematic sampling
50(3)
4.1.1 N/k is integer-valued
50(2)
4.1.2 N/k is not integer-valued
52(1)
4.2 Selection methods that guarantee fixed n
53(4)
4.2.1 Circular systematic sampling
53(1)
4.2.2 Fractional interval random start
54(3)
4.3 Estimation of sampling variance
57(5)
4.3.1 Biased estimation
57(1)
4.3.2 Unbiased estimation
58(4)
4.4 Unpredictable trend in sampling variance with n
62(1)
4.5 Warning: Pathological settings
62(1)
4.6 Nonresponse and oversampling
63(1)
4.7
Chapter comments
64(1)
Problems
65(3)
5 Stratified sampling 68(24)
5.1 Estimation of the population mean
69(3)
5.1.1 Expected value
69(1)
5.1.2 Sampling variance
70(1)
5.1.3 Numerical examples
70(2)
5.2 Estimation of the population proportion
72(1)
5.3 Estimation of the population total
72(1)
5.4 Estimation of sampling variance
73(1)
5.5 Allocation of the sample across strata
74(5)
5.5.1 Optimal allocation: Graphical analysis
75(1)
5.5.2 Optimal allocation: Analytical analysis
75(3)
5.5.3 Comments on optimal allocation
78(1)
5.6 Sample size determination
79(1)
5.7 Relative efficiency
80(3)
5.7.1 Proportional allocation
80(1)
5.7.2 Estimation of finite population variance
81(2)
5.8 Effective degrees of freedom
83(1)
5.9 Post-stratification
84(3)
5.9.1 Unconditional sampling variance
85(1)
5.9.2 Conditional sampling variance
86(1)
5.10
Chapter comments
87(1)
Problems
88(4)
6 Single-stage cluster sampling: Clusters of equal size 92(12)
6.1 Estimation of the population mean
93(1)
6.2 Sampling variance
94(3)
6.2.1 ANOVA/mean squares approach
95(1)
6.2.2 Intracluster correlation approach
96(1)
6.3 Estimation of the population total and proportion
97(1)
6.4 Estimation of sampling variance
98(1)
6.5 Estimation of finite population variance
99(1)
6.6 Sample size determination
100(1)
6.7 Relative efficiency
100(1)
6.8
Chapter comments
101(1)
Problems
102(2)
7 Ratio and regression estimation 104(36)
7.1 Estimation of the mean and total
105(10)
7.1.1 Graphical representation
106(1)
7.1.2 Sample space illustration
107(2)
7.1.3 Bias
109(1)
7.1.4 Sampling variance
110(2)
7.1.5 Estimation of sampling variance
112(1)
7.1.6 Sample size determination
112(1)
7.1.7 Relative efficiency
113(2)
7.2 Ratio estimation of a proportion
115(2)
7.3 Ratio estimation with stratified sampling
117(1)
7.3.1 Combined estimator
117(1)
7.3.2 Separate estimator
117(1)
7.3.3 Choosing between combined and separate estimators
118(1)
7.4 A model-based perspective
118(11)
7.4.1 Estimation of model parameters
119(5)
7.4.2 Prediction of population parameters
124(2)
7.4.3 Prediction error
126(2)
7.4.4 Prediction variance estimators
128(1)
7.5 Monte Carlo performance evaluation
129(3)
7.6 Mark-recapture estimation
132(1)
7.7
Chapter comments
133(2)
Problems
135(5)
8 Unequal probability sampling 140(33)
8.1 Unbiased ratio estimator
141(2)
8.2 Sampling with replacement
143(2)
8.2.1 Hansen-Hurwitz estimator
143(1)
8.2.2 Unbiasedness
144(1)
8.2.3 Sampling variance and variance estimation
144(1)
8.3 Sampling without replacement
145(12)
8.3.1 Horvitz-Thompson estimator
146(1)
8.3.2 Unbiasedness
147(1)
8.3.3 Sampling variance and variance estimation
148(2)
8.3.4 Alternative selection methods
150(2)
8.3.5 Strategy performance comparisons
152(3)
8.3.6 Survey cost comparisons
155(2)
8.4 Sampling distribution
157(2)
8.5 Systematic sampling
159(2)
8.6 Generality of Horvitz-Thompson estimation
161(1)
8.7 Generalized Horvitz-Thompson estimation
162(2)
8.7.1 Variance, covariance, and correlation estimators
163(1)
8.7.2 Mean-per-unit, ratio, and regression estimators
163(1)
8.7.3 Performance of generalized Horvitz-Thompson estimators
163(1)
8.8 Poisson sampling
164(1)
8.9 Nonresponse and oversampling
165(2)
8.9.1 Hansen-Hurwitz estimator
165(1)
8.9.2 Horvitz-Thompson estimator
166(1)
8.10
Chapter comments
167(1)
Problems
168(5)
9 Multi-stage sampling 173(27)
9.1 Two-stage sampling: Clusters of equal size
174(11)
9.1.1 Estimation of the population mean
175(1)
9.1.2 Expectation
176(3)
9.1.3 Sampling variance and its estimation
179(2)
9.1.4 Optimal allocation
181(3)
9.1.5 Net relative efficiency
184(1)
9.2 Two-stage sampling: Clusters of unequal size
185(10)
9.2.1 Single-stage cluster sampling
185(2)
9.2.2 Two-stage estimation of the population mean and total
187(1)
9.2.3 Sampling variance and its estimation
187(7)
9.2.4 Optimal allocation
194(1)
9.3
Chapter comments
195(2)
9.3.1 Generality of the multi-stage framework
195(1)
9.3.2 Taking advantage of ecological understanding
196(1)
9.3.3 Implications for large-scale natural resource surveys
196(1)
Problems
197(3)
10 Multi-phase sampling 200(19)
10.1 Two-phase estimation of the population mean and total
201(8)
10.1.1 Estimators
202(1)
10.1.2 Sampling variance and its estimation
202(1)
10.1.3 Sample space illustration
203(2)
10.1.4 Optimal allocation
205(2)
10.1.5 Net relative efficiency
207(2)
10.2 Two-phase ratio estimation of a proportion
209(3)
10.3 Two-phase sampling with stratification
212(3)
10.3.1 Estimation of the population mean and total
212(1)
10.3.2 Sampling variance and its estimation
212(1)
10.3.3 Optimal allocation
213(2)
10.3.4 Net relative efficiency
215(1)
10.4
Chapter comments
215(1)
Problems
216(3)
11 Adaptive sampling 219(21)
11.1 Adaptive cluster sampling
220(11)
11.1.1 Basic scheme
220(1)
11.1.2 Definitions
221(1)
11.1.3 Inclusion probabilities and expected sample size
222(2)
11.1.4 Estimators and relative efficiency
224(7)
11.2 Other adaptive sampling designs
231(5)
11.2.1 Single-stage strip and systematic designs
231(1)
11.2.2 Two-stage complete allocation cluster sampling
232(4)
11.3
Chapter comments
236(2)
Problems
238(2)
12 Spatially balanced sampling 240(29)
12.1 Introduction
240(4)
12.2 Finite populations
244(19)
12.2.1 Generalized random tessellation stratified sampling
244(7)
12.2.2 Balanced acceptance sampling
251(9)
12.2.3 Estimation
260(3)
12.3 Infinite populations
263(3)
12.3.1 Generalized random tessellation stratified sampling
263(1)
12.3.2 Balanced acceptance sampling
264(1)
12.3.3 Estimation
265(1)
12.4
Chapter comments
266(3)
13 Sampling through time 269(26)
13.1 Sampling on two occasions
270(5)
13.1.1 Design 1: Full retention of units across occasions
271(1)
13.1.2 Design 2: Independent SRS on each occasion
272(1)
13.1.3 Comparison of full retention and independent SRS designs
272(1)
13.1.4 Design 3: Partial retention/partial replacement
273(2)
13.2 Monitoring design
275(7)
13.2.1 Membership design
276(1)
13.2.2 Revisit design
277(5)
13.3 Estimation of status and trend
282(4)
13.3.1 Design-based estimation
283(1)
13.3.2 Estimators for some specific designs
284(2)
13.4 Sample size determination
286(2)
13.5 Dual frame sampling
288(2)
13.6
Chapter comments
290(5)
A Mathematical foundations 295(34)
A.1 Counting techniques
296(3)
A.1.1 Permutations
296(1)
A.1.2 k-Permutations
296(2)
A.1.3 Combinations
298(1)
A.1.4 Partitions
298(1)
A.2 Basic principles of probability theory
299(4)
A.2.1 Random experiment
299(1)
A.2.2 Sample space
300(1)
A.2.3 Outcome probability
300(1)
A.2.4 Event
301(1)
A.2.5 Event relations
301(1)
A.2.6 Union and intersection
302(1)
A.2.7 Conditional probability
302(1)
A.2.8 Other probability relations
302(1)
A.3 Discrete random variables
303(9)
A.3.1 Definition
303(1)
A.3.2 Probability distributions
303(2)
A.3.3 Probability relations
305(1)
A.3.4 Expectation
305(2)
A.3.5 Variance and coefficient of variation
307(2)
A.3.6 Covariance and correlation
309(1)
A.3.7 Total expectation and variance
310(1)
A.3.8 Indicator variables
311(1)
A.4 Key discrete probability distributions
312(5)
A.4.1 Uniform
312(1)
A.4.2 Bernoulli
313(1)
A.4.3 Multinoulli
313(1)
A.4.4 Binomial
314(1)
A.4.5 Multinomial
315(1)
A.4.6 Hypergeometric
316(1)
A.4.7 Multivariate hypergeometric
317(1)
A.5 Population variables
317(2)
A.5.1 Definition
317(1)
A.5.2 Population parameters
318(1)
A.6 Probability sampling
319(2)
A.6.1 Sampling experiment
319(1)
A.6.2 Sampling designs
320(1)
A.6.3 Inclusion probabilities
320(1)
A.6.4 Inclusion indicator variables
320(1)
A.7 Estimators
321(4)
A.7.1 Definition
321(1)
A.7.2 Sampling distribution
322(1)
A.7.3 Statistical properties
323(1)
A.7.4 Confidence intervals
324(1)
A.8 Delta method
325(2)
A.9 Lagrange multipliers
327(2)
References 329(8)
Index 337
David G. Hankin is a quantitative fisheries scientist who held a faculty position in the Department of Fisheries Biology at Humboldt State University (HSU) from 1978-2014. He holds a PhD in Fisheries Science from Cornell University. Over his long academic career at HSU, he taught introductory courses on fish population dynamics and sampling theory to audiences of senior and MS-level graduate students from natural resource and biological sciences. His fisheries research has focused on population dynamics and management of commercially important species, specifically Dungeness crab and Chinook salmon, with a special interest in life history and fishery inferences based on analysis of code wire tag recovery data. He has actively participated in the fishery management process, serving as a member of the Scientific and Statistical Committee of the Pacific Fisheries Management Council (1987-1992) and as a US member of the Committee on Scientific Cooperation (2001-2015).



Michael S. Mohr is Supervisory Mathematical Statistician and Fisheries Assessment Program Leader for NOAA Fisheries, Southwest Fisheries Science Center, Fisheries Ecology Division, and a Research Fellow at the Institute of Marine Sciences, University of California, Santa Cruz. He holds a Bachelor of Arts degree in Mathematics and a Master of Science degree in Fisheries Science from Humboldt State University and received post-graduate training in Statistics at the University of California, Berkeley. He has over 30 years of experience in quantitative population assessment, applying sampling theory to resource management problems, and developing statistical methods and estimators.



Ken B. Newman is principal researcher in statistical methodology for Biomathematics & Statistics Scotland and reader in the School of Mathematics, University of Edinburgh. He has a PhD in statistics from the University of Washington. From 1993 through 2003 he was on the faculty of the Division of Statistics at the University of Idaho and then moved to the University of St Andrews as a senior lecturer with a joint appointment in the Centre for Ecological and Environmental Modelling. Between 2006 and 2017, he was a mathematical statistician for the US Fish and Wildlife Service where he developed sampling designs for aquatic species and methods for estimating fish abundances and modeling fish population dynamics. His current position includes developing and applying statistical methods for agricultural and environmental processes as well as teaching Bayesian statistics.