Авторизация
Поиск по указателям
Efron B. — Large-Scale Inference. Empirical Bayes Methods for Estimation, Testing, and Prediction
Обсудите книгу на научном форуме
Нашли опечатку? Выделите ее мышкой и нажмите Ctrl+Enter
Название: Large-Scale Inference. Empirical Bayes Methods for Estimation, Testing, and Prediction
Автор: Efron B.
Аннотация: We live in a new age for statistical inference, where modern scientific technology such as microarrays and fMRI machines routinely produce thousands and sometimes millions of parallel data sets, each with its own estimation or testing problem. Doing thousands of problems at once involves more than repeated application of classical methods.
Taking an empirical Bayes approach, Bradley Efron, inventor of the bootstrap, shows how information accrues across problems in a way that combines Bayesian and frequentist ideas. Estimation, testing, and prediction blend in this framework, producing opportunities for new methodologies of increased power. New difficulties also arise, easily leading to flawed inferences.
This book takes a careful look at both the promise and pitfalls of large-scale statistical inference, with particular attention to false discovery rates, the most successful of the new statistical techniques. Emphasis is on the inferential ideas underlying technical developments, illustrated using a large number of real examples.
Язык:
Рубрика: Математика /
Статус предметного указателя: Готов указатель с номерами страниц
ed2k: ed2k stats
Год издания: 2010
Количество страниц: 263
Добавлена в каталог: 01.04.2012
Операции: Положить на полку |
Скопировать ссылку для форума | Скопировать ID
Предметный указатель
21st century xi 1 15 240
Acceleration 139
Accuracy calculations 27 126 141
Accuracy formula 114 139 148
Alpha 146—148 250
Analysis 163—165 184
Apparent error 216
atom 217 228 229
batting averages 7 8
BAX gene 163 164
Bayes x 4 5 10 12 14 18 19 23—25 28 55 57 69 82 90 95 97 192 205 217 220 223 231
Bayes factor 45 74
Bayes intervals 233
Bayes methods 2
Bayes rule 2 3 7 13 18 53 189 211 212 230 232 233
Bayes theorem 2 27 80 188 199 217
Bayes, hierarchical 13
Bayes, objective 13
Benjamini and Hochberg x 18 23 28 46 48 52 69
Benjamini — Hochberg algorithm 48 49 54 56 57 59 64 66 69 173 204
Bin centers 116 127 129
Bin width 79 90 116 118 119
Binomial distribution 24 31 48
Bivariate normal density 117
Block statistic 150
Block test 152
Bonferroni bound 17 35 36 38 39 43 81 202
Bonferroni’s method 81 200
Boole’s inequality 35 40
Bootstrap 113 115 138—140
Bootstrap methods 112 138
Bootstrap, non-parametric 115 138 139
Brain tumor study 207
BRCA data 152 162 167 168
calculations 153 180
Canonical parameter 217 243
Cardio data 149—151 153 162 249
Central counts 98 160 161
Central matching 97 99—101 104 105 108 112 129—131 226 227
Central matching, estimates 98—100 103 104 129 130 226
Chi-square data 92 93 107 112 206 249
Chi-square distribution 4 93
Chromosome 147 148 202—205
Chromosome, sites 202 203
Clemente 7—9
Coefficient of variation 23 24 77 128 178 236
Column permutation 153 165 166 168
Combination 28 120 163 184—186 192 209
Comish — Fisher 59 135 139
Comparability 94 185 206—209
Conditional probability 2 188 197 228
Confidence interval 12 14 24 230—233
Control algorithm 48 173 204 230 231 233
Corbet, Alexander 233—237 241
Correlation x 24 26 27 51 55 56 68 77 104 106 111 114 115 117 118 128 129 141 143 146—150 152 153 155 157 159—162 165 167 174 176—178 183 220 222 223
Correlation across cases 106 108 111 113
Correlation across units 106 109 110
Correlation clusters 68
Correlation corrections 221
Correlation distribution 118 120
Correlation effects 109 139 151 159 162 176
Correlation matrix 113 117 144 149 152 222
Correlation of t-values 157
Correlation structure 42 56 174 183 221
Correlation, column 139 143—146 152 153 156 168
Correlation, count 159
Correlation, ecological 149
correlation, negative 68 111 160 162
Correlation, rms 55 56 104 106 113 122 123 141 145—149 153—157 159 176 250
Correlation, row 111 143—147 153 156 168
Correlation, row and column 141—143
Correlation, sample 67 142—144 222
Correlation, simultaneous tests of 64 67
Count vector 116 117 126 159 161 196
Covariance matrix 113 116—119 125 128 142 148 149 154 155 157 167 221 236 246
Covariate 11 80 95 107 133 147 197 198 202 206
Covariate, unobserved 107 108 111
CRAN 174 215
Cross-validation 215 216 219 221—223 240 250
Cumulant 218 230 244
Cumulant, generating function 243 244
Cumulative distribution functions (cdf) 16 18 19 32 39 41 43 51 53 54 60 68 72 91 93 105 111 113—115 119 125 132—135 138 165 169 173 237
Cumulative distribution functions (cdf), empirical 53 55 124 234
Cumulative distribution functions (cdf), non-null 86
Deconvolution 218
Delta-method 23 115 126 127 130
Demeaning 154 155
Deviance function 244 246
Dichotomous case 212
Diffusion tensor imaging 26
Dirichlet 201 245
Disease status 202
distribution 64 109 110 184
DTI data 26 28 32 34 54 57 59—61 71—73 181 182 185 186 189 199 249
Dyslexic 26 34 181 183
Dyslexic and normal children 26 193
eBay 215 218—223 240 250
EER 43 45
effect size 181 212—214 217—220 223—225 227—231 233 250
Effect size, estimation 211 212 227 230 241
Effective sample size 153 155—157
Efficiency 1 27 90 195
Eigenvalues 142—144
Eigenvector 149 151 152
Empirical Bayes ix x 1 4 5 7 10 12—14 18 20—22 27—29 37 46 51—53 55 57 63 66 69 82 88—90 114 115 190 191 198 199 212 213 217—220 223 229 231 233—235 240 241 250
Empirical Bayes, confidence intervals 12 14 233
Empirical Bayes, methods 7 17 30 37 70 113 206 209 240
Empirical Bayes, posterior band 229
Empirical distribution 20 79 109 139 169
Empirical null 84 90 91 94—98 106—108 111 112 130 141 163 167 169 183 184 186 194 224—226 250
Empirical null, density 90 164
Empirical null, distribution 56 89 160 226 227
Empirical null, estimation 97 102 194 196
Empirical probability distribution 29
Enrichment 82 163 164 167 170 171 173 174 178 181—185 191 192 194
Estimation x xi 1—3 10 13 14 20 27 28 57 58 60 71 75—77 80 82 88—90 98 99 103 104 123 156 188 195 201 220 229 233 241 243
Estimation accuracy 113 148
Estimation, effect size 211 212 227 230 241
Estimation, empirical Bayes 1 4 13 63 113 240
Estimation, empirical null 97 102 194 196
Estimation, James — Stein 10 11 209
Estimation, maximum likelihood x 1 74 75 238 245 246
Exchangeability 80
Expected false and true positives 82 83
Exponential family xi 74 82 103 104 171 174 217 224 230 243—247
Exponential family, multiparameter 105 137 245 247
Exponential family, one-parameter 137 241 243
Exponential tilting 243 244
Expression matrix 47 61 133 144 146 149
False coverage rate 230—232
False discoveries 21 22 47 51 59 232
False discovery proportion 22 23 29 47—49 51 53 55 56 106 107 130 160 201
False discovery rate x xi 18 20 22 23 26 28 30 37 42 43 45—17 50 53 57 58 60 62 64 69 70 72 73 81—83 97 102 105 106 128 159 160 164 184—186 188 194 197—201 205 208 212 223 225
False discovery rate, Bayes 18 19 23—25 29 55 200
False discovery rate, local 19 28 29 57 70 73 77 78 82 83 86—88 90 97 183 185 187 197 223 226 229 250
False discovery rate, positive 51
False negative rate 57 69
FCR control 230 231 233
FDR control 30 37 45 46 48 50 51 57 58 69 70 173 204 231 233
Filtration 109
Fisher, R.A. ix x 14 15 31 32 36 45 48 65 111 112 234 237—239 241
Fisher’s linear discriminant function 211 212 214
Fisher’s prediction formula 237
Fisher’s scale of evidence 31 46
Fold-change 207—209
Frequentist ix x 1 2 5 10 14 15 18 30 37 48 54 115 199 200 205 230 233
Frequentist, methods 205 233
FWER 30 34 35 43 44 47 49 50 81 83 185 199 200 205
FWER, control 35 37 40 43
Gamma density 237
Gamma distribution 134
Gene expression 15 141 149 163 167 212
Gene set 163—171 173 174 176 181 183 184
Generalized linear model (GLM) 75 127 247
Generalized linear model (GLM), Poisson 131 247 248
GSEA statistic 165 172—174
Hermite polynomial 118 119
Hierarchical model 241
Hierarchical model, Bayes 27 28 96 182
Hierarchical model, normal 217 223—226 228 241
Hierarchical model, Poisson 240
High-dimensional statistical inference 1
Higher criticism 59 69 205
Hippocampus 197
HIV data 94 96 99 105 110—112 148 249
Holm’s procedure 36—38 40 81
Homogeneity assumption 123—125
Hyperparameter 13
Independence 24—26 36 39 50 55 56 59 66 76 84 92 106 111 113 124 125 128 130 141 152 153 221 222 235 236
Independence, assumption 24 26 55 60 106
Independence, column-wise 141 149 153 162
Indirect evidence 7 229 240
Influence function 127 129 130 139
James — Stein estimator 1 5—7 14 27
Jensen’s Inequality 22 23 155
K-FWER 43—45
Kidney data 11 14 249
Kolmogorov — Smimov test statistic 165 172
Kronecker product 154
Kullback — Leibler distance 244
kurtosis 132 135
Large-scale inference 3 9 34 163
Learning from the experience of others 10 27 184 206 229 240
Least squares 10 12 110 129 131 245
Lehmann alternatives 20
Leukemia data 63 64 98 111—114 124 132 138 146 147 224 225 249
Likelihood, x 1—3 5 13 74 75 102—104 171 238 245 246
Limma 184
Lindsey’s method 74—76 88 247
Linear logistic model 194 195
Linear regression 5 11 34 150 152 240
Local averaging 181—183
Local true discovery rate 84
Locfdr 34 75 76 82 86 101 103 104 127 153 226 250
Logistic regression estimate 189
Malaysian butterfly data 233 234 236
Marginal density 18 19 123 187 188 199 217 218 224 233
Marginal distribution 2 4 177
Max-T 40—43
Maximum likelihood estimator (MLE) 3 5 102
Maxmean statistic 171—173 184
Mean 2 8 9 24 28 56 68 77 87 98 104 107 115—117 121 125 126 132—135 137 141—144 147 149 154 156—159 161 165—168 172 173 177 178 184 211 214 250
Mean and variance 6 23 60 82 114 144 158 159 161 180 228 230 244
Mean, posterior 13 218 223
Mehler’s identity 118 121
Michigan lung cancer study 222 223 241
Microarray ix 1 15 27 28 43 45 61 81 91 96 106 107 109 121 141 143 145 148—150 152 163 164 167 181 206—208 212 215
Microarray experiment 61 107 112 133 143 152 163 223
Microarray study 27 61 64 91 109 113 148 163
Miller, Rupert xi 30 45 186 205
MIN 40 41
Missing mass problem 239
Missing species problem 233 237 241
Mixture density 18 21 29 71 74 88 96 101 114 187 250
MLE method 1 90 99 101 102 104 105 108 112 153 183 189
Multi-class model 117 119 187
multinomial 118 125 161 236 245 247
Multivariate normal 2 6 28 36 146 211 245
Natural parameter 137 217 243 244 247
Natural spline basis 76 90
Nearest shrunken centroid 215 240
Negative binomial 237 238
New species 234 235 237 238
Non-central 1 178 180
Non-central t 132 137 221
Non-central t, distribution 133
Non-null counts 84 85
Non-null counts, smoothed 84 85 96 124
Normal distribution 3 16 84 139 153 213
Normal distribution, matrix normal 154
Normal distribution, multivariate normal 211
Null and alternative densities 29
Null hypothesis x 16 26 31 32 48 49 51 55 58 65 74 83 89 96 97 105 111 113 132 135—137 141 149—151 164 165 170—172 174 194 195 201 224 227 250
Null hypothesis, blurry 97
Null hypothesis, complete 40—42
Null hypothesis, enrichment 194
Null proportion 20 28 49 54 60 66 90 98
Odds ratio 2 211 227
Optimal discovery procedure 45
p-values 31—41 45 48—53 57 66 70 136 141 167 169 173 174 200 204
P-values, adjusted 34—37 39—42 202
P53 data 163 166 173 184 194 250
P53 data, P53_UP 194 195
Pamr 215—219 240
PCER 43
Penalty 5 50 76 201 204 205
Penalty, correlation 114 118 119 122 123 128
Permutation algorithms 39
Permutation calculations 40 97 109 110 177—180 183
Permutation distribution 41 62 64 89 111 169 180
Permutation methods 41 42 110—112 149 171 174
Permutation null 97 109 111 141 149
Point estimate 22
Poisson assumption 25 170
Poisson deviance 76 248
Poisson process 234
Poisson regression 71 75 218 219 245
Poisson regression, estimate 74 114 123 127 196 233
Poisson selection model 170 174
Poisson-independence 24 25 58 200
Police data 93 95 112 226 227 249
Positive regression dependence 50 69
Posterior distribution 2 3 12 229
Posterior interval 12 13 232
Posterior probability 18 53 57 74 95
Power 34 48 57 74 83 85—89 104 153 163 170—173
Power, diagnostics 83 87 88
Prediction xi 1 7—10 28 211—217 223 228 229 234 235 237—241
Prediction error 7 214 219—221
Prediction rule 97 211—213 216 217 219—221 223
Prediction rule, ideal 213
Prediction, computer-intensive 212
Prior density 2 27 95—97 228
Prior knowledge 80 88
Prior probability 17 70 83 187 211 213
Prostate data 15 17 21 24 27 41 42 44 47 50 57 60—63 68 71—73 78—80 82 84 85 88 148 151 184 185 207 208 212 214—216 219 221 224 225 229 249 250
R-test x 28 32 64—66 68 69 91 96 112 149 163 168 249
Randomization 164—167 169 174 176 179
Ras pathway 164 166 171
Relevance x 81 82 94 185 186 196—198 205 206 209
Relevance function 14 197 198 209
Restandardization 168 169 173 174 176 178 179 183 184
Restandardization, restandardized 168 173 184
Risk function 3
Rms approximation 121—123 125 161
Robbins, Herbert x 1 14 88 240 241
Row 165 167—171 174 177 183 184
SAM 61—64 94 110 174 209
Scaling properties 80 81
Score function 171 246
Scoring function 165 169 170 174
Secant 19
Second derivative matrix 246
Selection bias 215 230
Separate analyses 185 186 199 205
Separation 184—186 196 199 200 205 209
Shakespeare data 238 239 241 250
Shakespeare, Wm. 238—241
Shakespeare, Wm., canon 238 239
Sidak procedure 36 37
Реклама