Machine Learning and Big Data with Kdb+/Q (Hardcover)

· 제목 : Machine Learning and Big Data with Kdb+/Q (Hardcover) 
· 분류 : 외국도서 > 경제경영 > 금융/재정 > 금융공학
· ISBN : 9781119404750
· 쪽수 : 640쪽
· 출판일 : 2019-12-31

I Language Fundamentals 1

1 Fundamentals of the q Programming Language 2

1.1 The (Not So Very) First Steps in q 2

1.2 Atoms and Lists 4

1.2.1 Casting Types 9

1.3 Basic Language Constructs 12

1.3.1 Assigning, Equality and Matching 12

1.3.2 Arithmetic Operations and Right-to-Left Evaluation: Introduction to q Philosophy 14

1.4 Basic Operators 16

1.5 Difference Between Strings and Symbols 26

1.5.1 Enumeration 27

1.6 Matrices and Basic Linear Algebra in q 28

1.7 Launching the Session: Additional Options 31

1.8 Summary and How-Tos 33

2 Dictionaries and Tables: The q Fundamentals 35

2.1 Dictionary 35

2.2 Table 37

2.3 The Truth about Tables 41

2.4 Keyed Tables Are Dictionaries 43

2.5 From a Vector Language to an Algebraic Language 44

3 Functions 48

3.1 Namespace 50

3.1.0.1 .quantQ. Namespace 51

3.2 The Six Adverbs 51

3.2.1 Each 51

3.2.1.1 Each 52

3.2.1.2 Each-left \: 52

3.2.1.3 Each-right /: 53

3.2.1.4 Cross product /:\: 53

3.2.1.5 Each-both ' 54

3.2.2 Each-prior ': 56

3.2.3 Compose (’) 57

3.2.4 Over and Fold / 57

3.2.5 Scan 58

3.2.5.1 EMA: The Exponential Moving Average 59

3.2.6 Converge 60

3.2.6.1 Converge-repeat 60

3.2.6.2 Converge-iterate 61

3.3 Apply 62

3.3.1 @ (apply) 62

3.3.2 (apply) 62

3.4 Protected Evaluations 64

3.5 Vector Operations 65

3.5.1 Aggregators 66

3.5.1.1 Simple Aggregators 66

3.5.1.2 Weighted Aggregators 66

3.5.2 Uniform Functions 67

3.5.2.1 Running Functions 67

3.5.2.2 Window Functions 67

3.6 Convention for User-Defined Functions 68

4 Editors and Other Tools 69

4.1 Console 69

4.2 Jupyter Notebook 70

4.3 GUIs 72

4.3.1 qStudio 73

4.3.2 Q Insight Pad 75

4.4 IDEs: IntelliJ IDEA 77

4.5 Conclusion 78

5 Debugging q Code 80

5.1 Introduction to Making It Wrong: Errors 80

5.1.1 Syntax Errors 81

5.1.2 Runtime Errors 81

5.1.2.1 The type error 82

5.1.2.2 Other Errors 84

5.2 Debugging the Code 86

5.3 Debugging Server-Side 88

II Data Operations 90

6 Splayed and Partitioned Tables 91

6.1 Introduction 91

6.2 Splayed Tables 93

6.3 Partitioned Tables 96

6.4 Conclusion 102

7 Joins 104

7.1 Comma operator 104

7.2 Join Functions 107

7.2.1 ij 107

7.2.2 ej 108

7.2.3 lj 109

7.2.4 pj 109

7.2.5 upsert 110

7.2.6 uj 111

7.2.7 aj 113

7.2.8 aj0 115

7.2.8.1 The next valid join 116

7.2.9 asof 119

7.2.10 wj 120

7.3 Advanced Example: Running TWAP 124

8 Parallelisation 130

8.1 Parallel Vector Operations 131

8.2 Parallelisation over Processes 133

8.3 Map Reduce 134

8.4 Advanced Topic: Parallel File/Directory Access 136

9 Data Cleaning and Filtering 139

9.1 Predicate Filtering 139

9.1.0.1 The where clause 139

9.1.0.2 Aggregation Filtering 141

9.2 Data Cleaning, Normalising and APIs 141

10 Parse Trees 143

10.1 Definition 144

10.1.1 Evaluation 144

10.1.2 Parse Tree Creation 147

10.1.3 Read-Only Evaluation 148

10.2 Functional Queries 148

10.2.1 Functional Select 151

10.2.2 Functional Exec 154

10.2.3 Functional Update 155

10.2.4 Functional Delete 156

11 A Few Use Cases 158

11.1 Rolling VWAP 158

11.1.1 N Tick VWAP 158

11.1.2 Time Window VWAP 159

11.2 Weighted Mid for N Levels of an Orderbook 160

11.3 Consecutive Runs of a Rule 161

11.4 Real-time Signals and Alerts 162

III Data Science 165

12 Basic Overview of Statistics 166

12.1 Histogram 166

12.2 First Moments 170

12.3 Testing Hypothesis 172

12.3.1 Normal p-values 172

12.3.2 Correlation 174

12.3.2.1 Implementation 175

12.3.3 t-test: One-sample 176

12.3.3.1 Implementation 177

12.3.4 t-test: Two-sample 178

12.3.4.1 Implementation 179

12.3.5 Sign Test 179

12.3.5.1 Implementation of the test 181

12.3.5.2 Median Test 184

12.3.6 Wilcoxon Signed-Rank Test 185

12.3.7 Rank Correlation and Somers’ D 187

12.3.7.1 Implementation 188

12.3.8 Multiple Hypothesis Testing 192

12.3.8.1 Bonferroni Correction 195

12.3.8.2 Šidák’s Correction 196

12.3.8.3 Holm’s Method 196

12.3.8.4 Example 197

13 Linear Regression 199

13.1 Linear Regression 200

13.2 Ordinary Least Squares 201

13.3 The Geometric Representation of Linear Regression 204

13.3.1 Moore–Penrose Pseudoinverse 205

13.3.2 Adding Intercept 207

13.4 Implementation of the OLS 210

13.5 Significance of Parameters 212

13.6 How Good is the Fit: R² 213

13.6.1 Adjusted R-squared 216

13.7 Relationship with Maximum Likelihood Estimation and AIC with Small Sample Correction 217

13.8 Estimation Suite 220

13.9 Comparing Two Nested Models: Towards a Stopping Rule 222

13.9.1 Comparing Two General Models 224

13.10 In/Out of Sample Operations 224

13.11Cross-validation 229

13.12 Conclusion 230

14 Time Series Econometrics 231

14.1 Autoregressive and Moving Average Processes 231

14.1.1 Introduction 231

14.1.2 AR(p) process 232

14.1.2.1 Simulation 232

14.1.2.2 Estimation of AR(p) Parameters 233

14.1.2.3 Least Square Method 234

14.1.2.4 Example 235

14.1.2.5 Maximum Likelihood Estimator 235

14.1.2.6 Yule-Walker Technique 235

14.1.3 MA(q) Process 237

14.1.3.1 Estimation of MA(q) Parameters 237

14.1.3.2 Simulation 238

14.1.3.3 Example 238

14.1.4 ARMA(p, q) Process 239

14.1.4.1 Invertibility of the ARMA(p, q) Process 239

14.1.4.2 Hannan-Rissanen Algorithm : Two-step Regression Estimation 240

14.1.4.3 Yule-Walker Estimation 240

14.1.4.4 Maximum Likelihood Estimation 240

14.1.4.5 Simulation 240

14.1.4.6 Forecast 241

14.1.5 ARIMA(p, d, q) Process 241

14.1.6 Code 242

14.1.6.1 Simulation 242

14.1.6.2 Estimation 243

14.1.6.3 Forecast 246

14.2 Stationarity and Granger Causality 249

14.2.1 Stationarity 249

14.2.2 Test of Stationarity - Dickey Fuller and Augmented Dickey Fuller tests 249

14.2.3 Granger Causality 250

14.3 Vector Autoregression 251

14.3.1 VAR(p) process 251

14.3.1.1 Notation 251

14.3.1.2 Estimator 252

14.3.1.3 Example 253

14.3.1.4 Code 256

14.3.2 VARX(p, q) process 259

14.3.2.1 Estimator 260

14.3.2.2 Code 260

15 Fourier Transform 262

15.1 Complex Numbers 262

15.1.1 Properties of Complex Numbers 263

15.2 Discrete Fourier Transform 269

15.3 Addendum: Quaternions 274

15.4 Addendum: Fractals 281

16 Eigensystem and PCA 284

16.1 Theory 284

16.2 Algorithms 286

16.2.1 QR Decomposition 287

16.2.2 QR Algorithm for Eigenvalues 289

16.2.3 Inverse Iteration 290

16.3 Implementation of Eigensystem Calculation 291

16.3.1 QR Decomposition 292

16.3.2 Inverse Iteration 296

16.4 The Data Matrix and the Principal Component Analysis 299

16.4.1 The Data Matrix 299

16.4.2 PCA: The First Principal Component 301

16.4.3 Second Principal Component 303

16.4.4 Terminology and Explained Variance 304

16.4.5 Dimensionality Reduction 306

16.4.6 PCA Regression (PCR) 307

16.5 Implementation of PCA 307

16.6 Appendix: Determinant 310

16.6.1 Theory 310

16.6.2 Techniques to Calculate Determinant 311

16.6.3 Implementation of the Determinant 312

17 Outlier Detection 315

17.1 Local Outlier Factor 316

18 Simulating Asset Prices 324

18.1 Stochastic Volatility Process with Price Jumps 324

18.2 Towards the Numerical Example 325

18.2.1 Numerical Preliminaries 326

18.2.2 Implementing Stochastic Volatility Process with Jumps 329

18.3 Conclusion 332

IV Machine Learning 333

19 Basic Principles of Machine Learning 334

19.1 Non-numeric Features and Normalisation 334

19.1.1 Non-numeric Features 334

19.1.1.1 Ordinal Features 335

19.1.1.2 Categorical Features 336

19.1.2 Normalisation 336

19.1.2.1 Normal Score 337

19.1.2.2 Range Scaling 338

19.2 Iteration: Constructing Machine Learning Algorithms 338

19.2.1 Iteration 338

19.2.2 Constructing Machine Learning Algorithms 341

20 Linear Regression with Regularisation 343

20.1 Bias-Variance Trade-off 343

20.2 Regularisation 345

20.3 Ridge Regression 346

20.4 Implementation of the Ridge Regression 348

20.4.1 Optimisation of the Regularisation Parameter 352

20.5 Lasso Regression 354

20.6 Implementation of the Lasso Regression 356

21 Nearest Neighbours 368

21.1 k-Nearest Neighbours Classifier 368

21.2 Prototype Clustering 372

21.3 Feature Selection: local Nearest Neighbours approach 377

21.3.0.1 Implementation 378

22 Neural Networks 384

22.1 Theoretical Introduction 384

22.1.1 Calibration 386

22.1.1.1 Backpropagation 387

22.1.2 The Learning Rate Parameter 389

22.1.3 Initialisation 390

22.1.4 Overfitting 390

22.1.5 Dimension of the Hidden Layer(s) 391

22.2 Implementation of Neural Networks 391

22.2.1 Multivariate Encoder 392

22.2.2 Neurons 393

22.2.3 Training the Neural Network 394

22.3 Examples 396

22.3.1 Binary Classification 396

22.3.2 M-class Classification 399

22.3.3 Regression 402

22.4 Possible Suggestions 407

23 AdaBoost with Stumps 408

23.1 Boosting 408

23.2 Decision Stumps 409

23.3 AdaBoost 410

23.4 Implementation of AdaBoost 411

23.5 Recommendation for Readers 417

24 Trees 418

24.1 Introduction to Trees 418

24.2 Regression Trees 420

24.2.1 Cost-Complexity Pruning 422

24.3 Classification Tree 423

24.4 Miscellaneous 425

24.5 Implementation of Trees 426

25 Forests 433

25.1 Bootstrap 433

25.2 Bagging 436

25.2.1 Out-of-bag 437

25.3 Implementation 438

25.3.1 Prediction 440

25.3.2 Feature Selection 442

26 Unsupervised Machine Learning: The Apriori Algorithm 445

26.1 Apriori Algorithm 446

26.2 Implementation of the Apriori Algorithm 447

27 Processing Information 456

27.1 Information Retrieval 456

27.1.1 Corpus: Leonardo da Vinci 456

27.1.2 Frequency Counting 457

27.1.3 tf-idf 461

27.2 Information as Features 465

27.2.1 Sample: Simulated Proteins 465

27.2.2 Kernels and Metrics for Proteins 466

27.2.3 Implementation of Inner Products and Nearest Neighbours Principles 467

27.2.4 Further Topics 470

28 Towards AI – Monte Carlo Tree Search 472

28.1 Multi-armed Bandit Problem 472

28.1.1 Analytic Solutions 474

28.1.2 Greedy algorithms 474

28.1.3 Confidence Based Algorithms 475

28.1.4 Bayesian Algorithms 476

28.1.5 Online Gradient Descent Algorithms 478

28.1.6 Implementation of Some Learning Algorithms 478

28.2 Monte Carlo Tree Search 487

28.2.1 Selection step 491

28.2.2 Expansion step 492

28.2.3 Simulation step 492

28.2.4 Back propagation step 493

28.2.5 Finishing the algorithm 493

28.2.6 Remarks and extensions 494

28.3 Monte Carlo Tree Search Implementation – Tic-tac-toe 495

28.3.1 Random Games 496

28.3.2 Towards the MCTS 499

28.3.3 Case Study 506

28.4 Monte Carlo Tree Search – Additional Comments 507

28.4.1 Policy and Value Networks 507

28.4.2 Reinforcement Learning 509

29 Econophysics: The Agent-Based Computational Models 510

29.1 Agent-based Modelling 511

29.1.1 Agent-based Models in Society 511

29.1.2 Agent-based Models in Finance 513

29.2 Ising Agent-Based Model for Financial Markets 514

29.2.1 Ising Model in Physics 514

29.2.2 Ising Model of Interacting Agents 514

29.2.3 Numerical Implementation 515

29.3 Conclusion 518

30 Epilogue: Art 520

Index 524

Bibliography 534