《R語言臨床預測模型實戰》是2023年清華大學出版社出版的圖書，作者是彭獻鎮。
基本介紹
- 中文名：R語言臨床預測模型實戰
- 作者：彭獻鎮
- 出版時間：2023年2月1日
- 出版社：清華大學出版社
- ISBN：9787302621119
- 定價：168 元
內容簡介,圖書目錄,作者簡介,
內容簡介
本書從R語言臨床預測模型基本原理講起，逐步深入到進階實戰，並配合實戰案例，重點介紹臨床預測模型的構建、評價、驗證，讓讀者可以快速領悟 3~5分預測模型 SCI（science citation index，科學引文索引）的套路與精髓，為後續衝擊10分以上SCI打基礎。
本書分為13章，主要內容有線性回歸、Logistic回歸、Cox回歸、競爭風險模型等；自變數篩選方法有傳統方法、逐步回歸法、Lasso法、隨機森林法、**子集法、主成分分析法等；模型可視化涉及多種形式的列線圖、Calibration校準曲線、ROC、DCA 曲線等圖形繪製，不僅涉及單模型的可視化，還涉及單模型多時點、多模型同時點的可視化；模型評價指標涉及C指數、AUC、NRI、IDI 等；模型驗證主要涉及簡單交叉驗證、K折交叉驗證、留一法交叉驗證及Bootstrap法。
本書內容通俗易懂，案例豐富，實用性強，適合R語言臨床預測模型的入門讀者和進階讀者，尤其是臨床、護理、公共衛生、藥學等專業的碩士研究生和博士研究生或從事相關研究的科研人員閱讀。另外，本書還可以作為相關培訓機構的教材。
圖書目錄
第1章 臨床預測模型概述 ············································1
1.1 如何構建預測模型 ·································································2
1.1.1 先單後多 ······················································································2
1.1.2 逐步回歸 ······················································································2
1.1.3 Lasso ···························································································2
1.1.4 隨機森林 ······················································································3
1.1.5 最優子集 ······················································································3
1.2 如何評價預測模型 ·································································3
1.2.1 擬合優度檢驗 ················································································3
1.2.2 ROC ····························································································4
1.2.3 Calibration ·····················································································4
1.2.4 DCA曲線 ·····················································································4
1.3 如何驗證預測模型 ·································································4
1.3.1 交叉驗證 ······················································································5
1.3.2 Bootstrap法 ···················································································5
1.4 小結 ···················································································5
第2章 線性回歸 ························································6
2.1 線性回歸概述 ·······································································7
2.1.1 相關的概念 ···················································································7
2.1.2 回歸的概念 ···················································································9
2.2 線性回歸實戰 ······································································11
2.2.1 簡單線性回歸 ···············································································11
IV
2.2.2 多重線性回歸 ···············································································17
2.3 小結 ··················································································25
第3章 Logistic回歸 ··················································27
3.1 概述 ··················································································28
3.1.1 二分類Logistic回歸的基本原理························································28
3.1.2 多分類Logistic回歸的基本原理························································30
3.1.3 有序Logistic回歸的基本原理···························································30
3.1.4 1∶m匹配條件Logistic回歸的基本原理···········································31
3.2 Logistic回歸分析實戰 ···························································31
3.2.1 二分類Logistic回歸 ·······································································31
3.2.2 多分類Logistic回歸 ·······································································41
3.2.3 有序Logistic ················································································49
3.2.4 1∶m匹配條件Logistic ·································································53
3.3 小結 ··················································································56
第4章生存資料分析 ················································59
4.1 概述 ··················································································60
4.1.1 Kaplan-Meier生存曲線的基本原理····················································60
4.1.2 Cox比例風險模型的基本原理 ··························································61
4.2 生存資料分析實戰 ································································62
4.2.1 Kaplan-Meier生存曲線 ···································································62
4.2.2 Cox比例風險模型 ·········································································72
4.3 小結 ··················································································81
第5章競爭風險模型 ················································83
5.1 概述 ··················································································84
5.2 競爭風險模型實戰 ································································85
5.2.1 Fine-Gray檢驗（單因素分析） ·························································85
5.2.2 crr（ ）函式法（多因素分析） ·····························································95
V
目錄
5.3 小結 ··················································································97
第6章 自變數篩選 ···················································99
6.1 傳統方法 ··········································································100
6.1.1 基於連續性資料 ··········································································100
6.1.2 基於二分類資料 ··········································································105
6.1.3 基於生存資料 ·············································································112
6.2 逐步法 ·············································································116
6.2.1 基於連續性資料 ··········································································117
6.2.2 基於二分類資料 ··········································································121
6.2.3 基於多分類資料 ··········································································124
6.2.4 基於有序資料 ·············································································126
6.2.5 基於1∶m匹配資料 ···································································128
6.2.6 基於生存資料 ·············································································130
6.2.7 基於競爭風險資料 ·······································································131
6.3 Lasso法 ···········································································134
6.3.1 基於連續性資料 ··········································································135
6.3.2 基於二分類資料 ··········································································144
6.3.3 基於多分類資料 ··········································································151
6.3.4 基於生存資料 ·············································································160
6.4 隨機森林法 ·······································································167
6.4.1 基於連續性資料 ··········································································168
6.4.2 基於二分類資料 ··········································································172
6.4.3 基於多分類資料 ··········································································177
6.4.4 基於生存資料 ·············································································180
6.4.5 基於競爭風險資料 ·······································································187
6.5 最優子集法 ·······································································191
6.5.1 基於連續性資料 ··········································································192
6.5.2 基於二分類資料 ··········································································193
6.5.3 基於多分類資料 ··········································································193
6.5.4 基於有序資料 ·············································································196
6.5.5 基於生存資料 ·············································································196
VI
6.5.6 基於競爭風險資料 ·······································································198
6.6 主成分分析法 ····································································200
6.7 小結 ················································································204
第7章列線圖·························································205
7.1 列線圖簡介 ·······································································206
7.2 基於連續資料 ····································································206
7.2.1 普通靜態列線圖 ··········································································206
7.2.2 網頁動態列線圖 ··········································································210
7.2.3 花樣列線圖 ················································································212
7.2.4 靜態列線圖的變種 ·······································································215
7.3 基於二分類資料 ·································································216
7.3.1 普通靜態列線圖 ··········································································216
7.3.2 網頁動態列線圖 ··········································································222
7.3.3 花樣列線圖 ················································································223
7.3.4 靜態列線圖的變種 ·······································································227
7.4 基於有序資料 ····································································228
7.4.1 普通靜態列線圖 ··········································································228
7.4.2 花樣列線圖 ················································································233
7.4.3 靜態列線圖的變種 ·······································································236
7.5 基於生存資料 ····································································237
7.5.1 普通靜態列線圖 ··········································································237
7.5.2 網頁動態列線圖 ··········································································246
7.5.3 花樣列線圖 ················································································248
7.5.4 靜態列線圖的變種 ·······································································253
7.5.5 生存機率列線圖 ··········································································255
7.6 基於競爭風險模型 ······························································256
7.6.1 普通靜態列線圖 ··········································································256
7.6.2 靜態列線圖的變種 ·······································································264
7.6.3 生存機率列線圖 ··········································································265
7.6.4 花樣列線圖 ················································································266
7.7 小結 ················································································270
第8章 Calibration校準曲線 ·································271
8.1 Calibration校準曲線簡介 ·····················································272
8.2 基於二分類資料 ·································································272
8.2.1 單一模型校準曲線 ·······································································272
8.2.2 多模型校準曲線 ··········································································281
8.3 基於生存資料 ····································································284
8.3.1 單模型單時點校準曲線 ·································································285
8.3.2 單模型多時點校準曲線 ·································································292
8.3.3 多模型同時點校準曲線 ·································································294
8.4 基於競爭風險模型 ······························································299
8.4.1 單模型單時點校準曲線 ·································································299
8.4.2 多模型同時點校準曲線 ·································································301
8.5 小結 ················································································304
第9章 C指數計算 ···················································305
9.1 C指數簡介 ·······································································306
9.2 基於二分類資料 ·································································306
9.3 基於生存資料 ····································································310
9.4 基於競爭風險模型 ······························································315
9.5 小結 ················································································318
第10章 ROC曲線 ···················································319
10.1 ROC曲線簡介 ·································································320
10.2 基於二分類資料 ·······························································320
10.2.1 單一模型ROC曲線 ····································································320
10.2.2 多模型ROC曲線 ·······································································328
10.3 基於生存資料 ··································································332
10.3.1 單模型單時點ROC曲線 ······························································332
10.3.2 單模型多時點ROC曲線 ······························································336
10.3.3 多模型同時點ROC曲線 ······························································339
VIII
10.4 基於競爭風險模型····························································343
10.4.1 單模型單時點ROC曲線 ······························································343
10.4.2 單模型多時點ROC曲線 ······························································345
10.4.3 多模型同時點ROC曲線 ······························································348
10.5 小結 ··············································································349
第11章 DCA曲線 ···················································350
11.1 DCA曲線簡介 ·································································351
11.2 基於二分類資料·······························································351
11.2.1 單一模型DCA曲線····································································351
11.2.2 多模型DCA曲線·······································································356
11.3 基於生存資料··································································359
11.3.1 單模型單時點DCA曲線 ······························································359
11.3.2 單模型多時點DCA曲線 ······························································363
11.3.3 多模型同時點DCA曲線 ······························································365
11.4 小結 ··············································································368
第12章 NRI、IDI計算 ·············································369
12.1 NRI、IDI簡介 ·································································370
12.2 基於二分類資料 ·······························································370
12.3 基於生存資料 ··································································378
12.4 小結 ··············································································384
第13章交叉驗證及Bootstrap ·····································385
13.1 概述 ··············································································386
13.2 簡單交叉驗證 ··································································389
13.3 K折交叉驗證 ··································································391
13.4 留一法交叉驗證 ·······························································393
13.5 Bootstrap ·········································································395
13.6 小結 ··············································································397
作者簡介
彭獻鎮，南京醫科大學康達學院流行病與統計學系主任，研究方向為慢性病流行病學， 擔任江蘇省預防醫學會青年委員，在國內外發表20餘篇學術論文，主持各級科研課題7項； 擅長SPSS、Stata、SAS、GraphPad Prism、PASS、R、Amos 等多種軟體的統計分析建模工作， 實戰經驗豐富，創建大鵬統計工作室至今，已為5000餘名碩士研究生、博士研究生提供統計分析、科研設計服務。
