Muutke küpsiste eelistusi

E-raamat: Advances in Financial Machine Learning

4.13/5 (497 hinnangut Goodreads-ist)
  • Formaat: EPUB+DRM
  • Ilmumisaeg: 02-Feb-2018
  • Kirjastus: John Wiley & Sons Inc
  • Keel: eng
  • ISBN-13: 9781119482109
Teised raamatud teemal:
  • Formaat - EPUB+DRM
  • Hind: 40,95 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
Advances in Financial Machine LearningSuurem pilt
  • Formaat: EPUB+DRM
  • Ilmumisaeg: 02-Feb-2018
  • Kirjastus: John Wiley & Sons Inc
  • Keel: eng
  • ISBN-13: 9781119482109
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

Learn to understand and implement the latest machine learning innovations to improve your investment performance

Machine learning (ML) is changing virtually every aspect of our lives. Today, ML algorithms accomplish tasks that until recently only expert humans could perform. And finance is ripe for disruptive innovations that will transform how the following generations understand money and invest.

In the book, readers will learn how to:





Structure big data in a way that is amenable to ML algorithms Conduct research with ML algorithms on big data Use supercomputing methods and back test their discoveries while avoiding false positives

Advances in Financial Machine Learning addresses real life problems faced by practitioners every day, and explains scientifically sound solutions using math, supported by code and examples. Readers become active users who can test the proposed solutions in their individual setting.

Written by a recognized expert and portfolio manager, this book will equip investment professionals with the groundbreaking tools needed to succeed in modern finance.

Arvustused

"Advances in Financial Machine Learning is a very interesting book... the author knows his subject." (BCS: The Chartered Institute for IT, August 2018)

"Prado's book clearly illustrates how fast this world is moving, and how deep you need to dive if you are to excel and deliver top of the range solutions and above the curve performing algorithms." (Irish Tech News, July 2018)

About the Author xxi
Preamble 1(2)
1 Financial Machine Learning as a Distinct Subject
3(18)
1.1 Motivation
3(1)
1.2 The Main Reason Financial Machine Learning Projects Usually Fail
4(2)
1.2.1 The Sisyphus Paradigm
4(1)
1.2.2 The Meta-Strategy Paradigm
5(1)
1.3 Book Structure
6(6)
1.3.1 Structure by Production Chain
6(3)
1.3.2 Structure by Strategy Component
9(3)
1.3.3 Structure by Common Pitfall
12(1)
1.4 Target Audience
12(1)
1.5 Requisites
13(1)
1.6 FAQs
14(4)
1.7 Acknowledgments
18(3)
Exercises
19(1)
References
20(1)
Bibliography
20(1)
PART 1 DATA ANALYSIS
21(70)
2 Financial Data Structures
23(20)
2.1 Motivation
23(1)
2.2 Essential Types of Financial Data
23(2)
2.2.1 Fundamental Data
23(1)
2.2.2 Market Data
24(1)
2.2.3 Analytics
25(1)
2.2.4 Alternative Data
25(1)
2.3 Bars
25(7)
2.3.1 Standard Bars
26(3)
2.3.2 Information-Driven Bars
29(3)
2.4 Dealing with Multi-Product Series
32(6)
2.4.1 The ETF Trick
33(2)
2.4.2 PCA Weights
35(1)
2.4.3 Single Future Roll
36(2)
2.5 Sampling Features
38(5)
2.5.1 Sampling for Reduction
38(1)
2.5.2 Event-Based Sampling
38(2)
Exercises
40(1)
References
41(2)
3 Labeling
43(16)
3.1 Motivation
43(1)
3.2 The Fixed-Time Horizon Method
43(1)
3.3 Computing Dynamic Thresholds
44(1)
3.4 The Triple-Barrier Method
45(3)
3.5 Learning Side and Size
48(2)
3.6 Meta-Labeling
50(1)
3.7 How to Use Meta-Labeling
51(2)
3.8 The Quantamental Way
53(1)
3.9 Dropping Unnecessary Labels
54(5)
Exercises
55(1)
Bibliography
56(3)
4 Sample Weights
59(16)
4.1 Motivation
59(1)
4.2 Overlapping Outcomes
59(1)
4.3 Number of Concurrent Labels
60(1)
4.4 Average Uniqueness of a Label
61(1)
4.5 Bagging Classifiers and Uniqueness
62(6)
4.5.1 Sequential Bootstrap
63(1)
4.5.2 Implementation of Sequential Bootstrap
64(1)
4.5.3 A Numerical Example
65(1)
4.5.4 Monte Carlo Experiments
66(2)
4.6 Return Attribution
68(2)
4.7 Time Decay
70(1)
4.8 Class Weights
71(4)
Exercises
72(1)
References
73(1)
Bibliography
73(2)
5 Fractionally Differentiated Features
75(16)
5.1 Motivation
75(1)
5.2 The Stationarity vs. Memory Dilemma
75(1)
5.3 Literature Review
76(1)
5.4 The Method
77(3)
5.4.1 Long Memory
77(1)
5.4.2 Iterative Estimation
78(2)
5.4.3 Convergence
80(1)
5.5 Implementation
80(4)
5.5.1 Expanding Window
80(2)
5.5.2 Fixed-Width Window Fracdiff
82(2)
5.6 Stationarity with Maximum Memory Preservation
84(4)
5.7 Conclusion
88(3)
Exercises
88(1)
References
89(1)
Bibliography
89(2)
PART 2 MODELLING
91(48)
6 Ensemble Methods
93(10)
6.1 Motivation
93(1)
6.2 The Three Sources of Errors
93(1)
6.3 Bootstrap Aggregation
94(4)
6.3.1 Variance Reduction
94(2)
6.3.2 Improved Accuracy
96(1)
6.3.3 Observation Redundancy
97(1)
6.4 Random Forest
98(1)
6.5 Boosting
99(1)
6.6 Bagging vs. Boosting in Finance
100(1)
6.7 Bagging for Scalability
101(2)
Exercises
101(1)
References
102(1)
Bibliography
102(1)
7 Cross-Validation in Finance
103(10)
7.1 Motivation
103(1)
7.2 The Goal of Cross-Validation
103(1)
7.3 Why K-Fold CV Fails in Finance
104(1)
7.4 A Solution: Purged K-Fold CV
105(4)
7.4.1 Purging the Training Set
105(2)
7.4.2 Embargo
107(1)
7.4.3 The Purged K-Fold Class
108(1)
7.5 Bugs in Sklearn's Cross-Validation
109(4)
Exercises
110(1)
Bibliography
111(2)
8 Feature Importance
113(16)
8.1 Motivation
113(1)
8.2 The Importance of Feature Importance
113(1)
8.3 Feature Importance with Substitution Effects
114(3)
8.3.1 Mean Decrease Impurity
114(2)
8.3.2 Mean Decrease Accuracy
116(1)
8.4 Feature Importance without Substitution Effects
117(4)
8.4.1 Single Feature Importance
117(1)
8.4.2 Orthogonal Features
118(3)
8.5 Parallelized vs. Stacked Feature Importance
121(1)
8.6 Experiments with Synthetic Data
122(7)
Exercises
127(1)
References
127(2)
9 Hyper-Parameter Tuning with Cross-Validation
129(10)
9.1 Motivation
129(1)
9.2 Grid Search Cross-Validation
129(2)
9.3 Randomized Search Cross-Validation
131(3)
9.3.1 Log-Uniform Distribution
132(2)
9.4 Scoring and Hyper-parameter Tuning
134(5)
Exercises
135(1)
References
136(1)
Bibliography
137(2)
PART 3 BACKTESTING
139(108)
10 Bet Sizing
141(10)
10.1 Motivation
141(1)
10.2 Strategy-Independent Bet Sizing Approaches
141(1)
10.3 Bet Sizing from Predicted Probabilities
142(2)
10.4 Averaging Active Bets
144(1)
10.5 Size Discretization
144(1)
10.6 Dynamic Bet Sizes and Limit Prices
145(6)
Exercises
148(1)
References
149(1)
Bibliography
149(2)
11 The Dangers of Backtesting
151(10)
11.1 Motivation
151(1)
11.2 Mission Impossible: The Flawless Backtest
151(1)
11.3 Even If Your Backtest Is Flawless, It Is Probably Wrong
152(1)
11.4 Backtesting Is Not a Research Tool
153(1)
11.5 A Few General Recommendations
153(2)
11.6 Strategy Selection
155(6)
Exercises
158(1)
References
158(1)
Bibliography
159(2)
12 Backtesting through Cross-Validation
161(8)
12.1 Motivation
161(1)
12.2 The Walk-Forward Method
161(1)
12.2.1 Pitfalls of the Walk-Forward Method
162(1)
12.3 The Cross-Validation Method
162(1)
12.4 The Combinatorial Purged Cross-Validation Method
163(3)
12.4.1 Combinatorial Splits
164(1)
12.4.2 The Combinatorial Purged Cross-Validation Backtesting Algorithm
165(1)
12.4.3 A Few Examples
165(1)
12.5 How Combinatorial Purged Cross-Validation Addresses Backtest Overfitting
166(3)
Exercises
167(1)
References
168(1)
13 Backtesting on Synthetic Data
169(26)
13.1 Motivation
169(1)
13.2 Trading Rules
169(1)
13.3 The Problem
170(2)
13.4 Our Framework
172(1)
13.5 Numerical Determination of Optimal Trading Rules
173(3)
13.5.1 The Algorithm
173(1)
13.5.2 Implementation
174(2)
13.6 Experimental Results
176(16)
13.6.1 Cases with Zero Long-Run Equilibrium
177(3)
13.6.2 Cases with Positive Long-Run Equilibrium
180(2)
13.6.3 Cases with Negative Long-Run Equilibrium
182(10)
13.7 Conclusion
192(3)
Exercises
192(1)
References
193(2)
14 Backtest Statistics
195(16)
14.1 Motivation
195(1)
14.2 Types of Backtest Statistics
195(1)
14.3 General Characteristics
196(2)
14.4 Performance
198(1)
14.4.1 Time-Weighted Rate of Return
198(1)
14.5 Runs
199(3)
14.5.1 Returns Concentration
199(2)
14.5.2 Drawdown and Time under Water
201(1)
14.5.3 Runs Statistics for Performance Evaluation
201(1)
14.6 Implementation Shortfall
202(1)
14.7 Efficiency
203(3)
14.7.1 The Sharpe Ratio
203(1)
14.7.2 The Probabilistic Sharpe Ratio
203(1)
14.7.3 The Deflated Sharpe Ratio
204(1)
14.7.4 Efficiency Statistics
205(1)
14.8 Classification Scores
206(1)
14.9 Attribution
207(4)
Exercises
208(1)
References
209(1)
Bibliography
209(2)
15 Understanding Strategy Risk
211(10)
15.1 Motivation
211(1)
15.2 Symmetric Payouts
211(2)
15.3 Asymmetric Payouts
213(3)
15.4 The Probability of Strategy Failure
216(5)
15.4.1 Algorithm
217(1)
15.4.2 Implementation
217(2)
Exercises
219(1)
References
220(1)
16 Machine Learning Asset Allocation
221(26)
16.1 Motivation
221(1)
16.2 The Problem with Convex Portfolio Optimization
221(1)
16.3 Markowitz's Curse
222(1)
16.4 From Geometric to Hierarchical Relationships
223(8)
16.4.1 Tree Clustering
224(5)
16.4.2 Quasi-Diagonalization
229(1)
16.4.3 Recursive Bisection
229(2)
16.5 A Numerical Example
231(3)
16.6 Out-of-Sample Monte Carlo Simulations
234(2)
16.7 Further Research
236(2)
16.8 Conclusion
238(1)
Appendices
239(8)
16.A.1 Correlation-based Metric
239(1)
16.A.2 Inverse Variance Allocation
239(1)
16.A.3 Reproducing the Numerical Example
240(2)
16.A.4 Reproducing the Monte Carlo Experiment
242(2)
Exercises
244(1)
References
245(2)
PART 4 USEFUL FINANCIAL FEATURES
247(54)
17 Structural Breaks
249(14)
17.1 Motivation
249(1)
17.2 Types of Structural Break Tests
249(1)
17.3 CUSUM Tests
250(1)
17.3.1 Brown-Durbin-Evans CUSUM Test on Recursive Residuals
250(1)
17.3.2 Chu-Stinchcombe-White CUSUM Test on Levels
251(1)
17.4 Explosiveness Tests
251(12)
17.4.1 Chow-Type Dickey-Fuller Test
251(1)
17.4.2 Supremum Augmented Dickey-Fuller
252(7)
17.4.3 Sub- and Super-Martingale Tests
259(2)
Exercises
261(1)
References
261(2)
18 Entropy Features
263(18)
18.1 Motivation
263(1)
18.2 Shannon's Entropy
263(1)
18.3 The Plug-in (or Maximum Likelihood) Estimator
264(1)
18.4 Lempel-Ziv Estimators
265(4)
18.5 Encoding Schemes
269(2)
18.5.1 Binary Encoding
270(1)
18.5.2 Quantile Encoding
270(1)
18.5.3 Sigma Encoding
270(1)
18.6 Entropy of a Gaussian Process
271(1)
18.7 Entropy and the Generalized Mean
271(4)
18.8 A Few Financial Applications of Entropy
275(6)
18.8.1 Market Efficiency
275(1)
18.8.2 Maximum Entropy Generation
275(1)
18.8.3 Portfolio Concentration
275(1)
18.8.4 Market Microstructure
276(1)
Exercises
277(1)
References
278(1)
Bibliography
279(2)
19 Microstructural Features
281(20)
19.1 Motivation
281(1)
19.2 Review of the Literature
281(1)
19.3 First Generation: Price Sequences
282(4)
19.3.1 The Tick Rule
282(1)
19.3.2 The Roll Model
282(1)
19.3.3 High-Low Volatility Estimator
283(1)
19.3.4 Corwin and Schultz
284(2)
19.4 Second Generation: Strategic Trade Models
286(4)
19.4.1 Kyle's Lambda
286(2)
19.4.2 Amihud's Lambda
288(1)
19.4.3 Hasbrouck's Lambda
289(1)
19.5 Third Generation: Sequential Trade Models
290(3)
19.5.1 Probability of Information-based Trading
290(2)
19.5.2 Volume-Synchronized Probability of Informed Trading
292(1)
19.6 Additional Features from Microstructural Datasets
293(2)
19.6.1 Distibution of Order Sizes
293(1)
19.6.2 Cancellation Rates, Limit Orders, Market Orders
293(1)
19.6.3 Time-Weighted Average Price Execution Algorithms
294(1)
19.6.4 Options Markets
295(1)
19.6.5 Serial Correlation of Signed Order Flow
295(1)
19.7 What Is Microstructural Information?
295(6)
Exercises
296(2)
References
298(3)
PART 5 HIGH-PERFORMANCE COMPUTING RECIPES
301(52)
20 Multiprocessing and Vectorization
303(16)
20.1 Motivation
303(1)
20.2 Vectorization Example
303(1)
20.3 Single-Thread vs. Multithreading vs. Multiprocessing
304(2)
20.4 Atoms and Molecules
306(3)
20.4.1 Linear Partitions
306(1)
20.4.2 Two-Nested Loops Partitions
307(2)
20.5 Multiprocessing Engines
309(6)
20.5.1 Preparing the Jobs
309(2)
20.5.2 Asynchronous Calls
311(1)
20.5.3 Unwrapping the Callback
312(1)
20.5.4 Pickle/Unpickle Objects
313(1)
20.5.5 Output Reduction
313(2)
20.6 Multiprocessing Example
315(4)
Exercises
316(1)
Reference
317(1)
Bibliography
317(2)
21 Brute Force and Quantum Computers
319(10)
21.1 Motivation
319(1)
21.2 Combinatorial Optimization
319(1)
21.3 The Objective Function
320(1)
21.4 The Problem
321(1)
21.5 An Integer Optimization Approach
321(4)
21.5.1 Pigeonhole Partitions
321(2)
21.5.2 Feasible Static Solutions
323(1)
21.5.3 Evaluating Trajectories
323(2)
21.6 A Numerical Example
325(4)
21.6.1 Random Matrices
325(1)
21.6.2 Static Solution
326(1)
21.6.3 Dynamic Solution
327(1)
Exercises
327(1)
References
328(1)
22 High-Performance Computational Intelligence and Forecasting Technologies
329(24)
Kesheng Wu
Horst D. Simon
22.1 Motivation
329(1)
22.2 Regulatory Response to the Flash Crash of 2010
329(1)
22.3 Background
330(1)
22.4 HPC Hardware
331(4)
22.5 HPC Software
335(2)
22.5.1 Message Passing Interface
335(1)
22.5.2 Hierarchical Data Format 5
336(1)
22.5.3 In Situ Processing
336(1)
22.5.4 Convergence
337(1)
22.6 Use Cases
337(12)
22.6.1 Supernova Hunting
337(1)
22.6.2 Blobs in Fusion Plasma
338(2)
22.6.3 Intraday Peak Electricity Usage
340(1)
22.6.4 The Flash Crash of 2010
341(5)
22.6.5 Volume-synchronized Probability of Informed Trading Calibration
346(1)
22.6.6 Revealing High Frequency Events with Non-uniform Fast Fourier Transform
347(2)
22.7 Summary and Call for Participation
349(1)
22.8 Acknowledgments
350(3)
References
350(3)
Index 353
DR. MARCOS LÓPEZ DE PRADO is a principal at AQR Capital Management, and its head of machine learning. Marcos is also a research fellow at Lawrence Berkeley National Laboratory (U.S. Department of Energy, Office of Science). SSRN ranks him as one of the most-read authors in economics, and he has published dozens of scientific articles on machine learning and supercomputing in the leading academic journals. Marcos earned a PhD in financial economics (2003), a second PhD in mathematical finance (2011) from Universidad Complutense de Madrid, and is a recipient of Spain's National Award for Academic Excellence (1999). He completed his post-doctoral research at Harvard University and Cornell University, where he teaches a graduate course in financial machine learning at the School of Engineering. Marcos has an Erdös #2 and an Einstein #4 according to the American Mathematical Society.
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97811194821096e.html
Märksõnad: