Paczkowski Walter / Walter Paczkowski - Hands-On Prescriptive Analytics / Practical Prescriptive Analytics [2024, PDF/EPUB, ENG] torrent

Книги и журналы » Компьютерная литература » Программирование (книги)

Hands-On Prescriptive Analytics: Optimizing Your Decision Making with Python / Практическая предписывающая аналитика: Оптимизация процесса принятия решений с помощью Python

Год издания: 2024
Автор: Paczkowski Walter / Пачковски Вальтер

Издательство: O’Reilly Media, Inc.
ISBN: 978-1-098-15317-5
Язык: Английский

Формат: PDF/EPUB
Качество: Издательский макет или текст (eBook)
Интерактивное оглавление: Да
Количество страниц: 412

Описание: Business decisions in any context—operational, tactical, or strategic—can have considerable consequences. Whether the outcome is positive and rewarding or negative and damaging to the business, its employees, and stakeholders is unknown when action is approved. These decisions are usually made under the proverbial cloud of uncertainty.
With this practical guide, data analysts, data scientists, and business analysts will learn why and how maximizing positive consequences and minimizing negative ones requires three forms of rich information: Descriptive analytics explores the results from an action—what has already happened. Predictive analytics focuses on what could happen. The third, prescriptive analytics, informs us what should happen in the future.
While all three are important for decision-makers, the primary focus of this book is on the third: prescriptive analytics. Author Walter R. Paczkowski, Ph.D. shows you:
The distinction among descriptive, predictive, and prescriptive analytics
How predictive analytics produces a menu of action options
How prescriptive analytics narrows the menu of action options
The forms of prescriptive analytics: eight prescriptive methods
Two broad classes of these methods: non-stochastic and stochastic
How to develop prescriptive analyses for action recommendations
Ways to use an appropriate tool-set in Python

Деловые решения в любом контексте — оперативном, тактическом или стратегическом — могут иметь значительные последствия. Будет ли результат положительным и приносящим удовлетворение, или отрицательным и наносящим ущерб бизнесу, его сотрудникам и заинтересованным сторонам, неизвестно, когда будет принято решение. Такие решения обычно принимаются в условиях пресловутой неопределенности.
Из этого практического руководства аналитики данных, data scientists и бизнес—аналитики узнают, почему и как для максимизации положительных последствий и минимизации отрицательных требуются три вида обширной информации: Описательная аналитика исследует результаты действия - то, что уже произошло. Прогнозная аналитика фокусируется на том, что может произойти. Третья, предписывающая аналитика, информирует нас о том, что должно произойти в будущем.
Хотя все три аспекта важны для лиц, принимающих решения, основное внимание в этой книге уделяется третьему аспекту: предписывающей аналитике. Автор Уолтер Р. Пачковски, доктор философии, расскажет вам о том, как:
Различие между описательной, прогностической и предписывающей аналитикой
Как в predictive analytics создается меню вариантов действий
Как в предписывающей аналитике список вариантов действий сужается
Формы предписывающей аналитики: восемь предписывающих методов
Существует два основных класса этих методов: нестохастические и стохастические случайности
Как разработать предписывающий анализ для выработки рекомендаций к действию
Способы использования соответствующего набора инструментов в Python
[spoiler="Примеры страниц (скриншоты)"]

[/spoiler]
[spoiler="Оглавление"]
Preface ix
Part I. Introduction and Background
1. An Analytical Framework 3
A Decision-Making Framework 4
The Analytics Evolution 7
The Data Science Split 9
The Information Glue 10
How the Analytics Fit Together 13
Descriptive Analytics 15
Predictive Analytics 17
Uncertainty: Multiple Sources and Problems 27
Uncertainty Versus Risk 28
The Menu Cost of Uncertainty 29
Probabilities and Uncertainty 30
Prescriptive Analytics as a Separate Discipline 32
Decision-Makers’ Options 33
Business Goals and Constraints 36
Key Decision Question: What Is the Best Decision? 37
The Analytics Flow 37
Prescriptive Analytics and Decision Making 38
Summary 39
2. Prescriptive Methods: Overview 41
Introduction to Prescriptive Analytics Methods 41
Proprietary Methods 42
Public Domain Methods 42
Summary of Prescriptive Analytics Methods 43
Methods Reviewed in the Literature 44
Three Categories of Methods 52
Umbrella Classes: Non-Stochastic and Stochastic 57
Definition of Non-Stochastic and Stochastic Methods 57
Examples Based on Scale-Views 58
The Role of Operations Research 59
Summary 60
Part II. Essential Background Material
3. Python Essentials 63
Python Structure: Overview 64
Worldwide Community 64
Extensive Array of Packages 64
Easy to Use 65
Operating Systems 65
A Programming Language 66
It Is Free 67
Python Basics 67
Naming Conventions 72
Python Data Structures 72
Iterables 78
Basic Python Operators 78
Introduction to Built-In Functions 79
Introduction to User-Defined Functions 82
Conditional Statements: if-else 85
Python Looping Constructs 87
for loops 87
while loops 90
Python Packages 90
Data Management 91
Data Visualization 94
Statistical Analysis and Modeling 96
Working with Python Packages 100
Using Anaconda 100
Updating Python Packages 101
Installing Packages 101
Importing Packages 101
Go-To References 105
Summary 105
4. Probability Essentials 107
The World Is Ruled by Probabilities 108
What Are Probabilities? 110
Fundamental Probability Concepts 112
Frequency-Based Probabilities 112
Counting Functions 113
Independence and Conditional Probability 118
Summary of Probability Rules 123
Limit Definition of Probabilities 124
Subjective-Based Probabilities: Introduction 124
Bayes’ Theorem: Derivation 126
Bayes’ Theorem: Python Implementation 127
Probability Distributions: Overview 128
Three Basic Probability Distributions: Binomial, Uniform, and Normal 129
Key Distribution Parameters 133
Summary 135
Part III. Non-Stochastic Prescriptive Analytic Methods
5. Mathematical Programming: Overview 139
Background 139
Reason for Popularity 140
The Objective Function: Prescriptive Analytics Goal 140
Linear Programming 141
Technical Overview 141
Menus and Linear Programming 150
Python Use-Cases 151
Integer Programming 162
Technical Overview 163
Menus and Integer Programming 165
Python Use-Cases 165
Mixed Integer Programming 178
Technical Overview 178
Python Use-Case 178
Summary 179
6. Decision Tree Analysis: Overview 181
Extending the Menu into Time 182
Introduction to Decision Trees 183
Clarification of Decision Trees 184
Background 188
Use of Trees in Decisions 188
Comparing the Two Decision Trees 191
Python Use-Case 192
Use-Case Background 192
Use-Case Detailed Data 194
Role of the DADT 196
DADT Analysis 198
Reaching a Decision Using a DADT 203
Summary 205
DADT Function 205
Part IV. Stochastic Prescriptive Analytic Methods
7. Simulation Essentials 211
What Is a Simulation? 212
The Simulation Age 214
Types of Simulations 214
Non-Stochastic Simulations: The Process 216
Stochastic Simulations: The Process 216
Pseudo-Random Number Generators: A Brief Introduction 217
Simulation Models: Overview 224
KPMs to Measure 225
Aggregation Methods 225
The Need for Stochastic Simulations 226
The Extent of Uncertainty 226
Randomness and the Degree of Uncertainty 227
Summary 227
8. Simulation Examples 229
Example 1: Coin Toss 229
Example 2: Die Toss 231
Example 3: Regression Analysis 238
Example 4: Mathematical Programming 246
Example 5: Decision Tree 256
Summary 265
9. Developing Menu Options 267
The Nature of What-If Questions 268
Menu Generating Questions: A Deep Dive 269
The Structure of What-If Questions 271
What-If Analysis Versus Sensitivity Analysis 273
Non-Stochastic Use-Cases 275
Pricing What-If Analysis: Basic 276
Pricing What-If Analysis: Advanced 280
Stochastic Use-Case: Synthetic Data 292
Specifying the Process to Simulate 293
Example of a System’s Process Flow 294
Summary 310
10. Developing Menu Priors 311
Background 311
Digression on Beliefs and Priors 312
Developing Probability Weights 313
Eliciting Probability Distributions of Beliefs 314
Elicitation Method 1: Experimental Design-Based 315
Elicitation Method 2: Direct Questioning 319
Elicitation Method 3: Activities 321
Analyzing Elicited Probability Distributions 322
Elicitation Analysis 1: Averaging 322
Elicitation Analysis 2: CoDA 323
Elicitation Analysis 3: Bootstrapping 324
Python Use-Case 326
Elicitation Example Analysis 1: Averaging 328
Elicitation Example Analysis 2: CoDA 336
Elicitation Example Analysis 3: Bootstrapping 339
Summary 343
11. One-Time Decisions 345
Evidence of the Problem 345
Sequential Decisions: Introduction 347
Sequential Decisions 1: The Business Case 348
Sequential Decisions 2: Post-Business Case 352
Sequential Analysis: Advanced Framework 356
Markov Decision Problem 359
Simulations and Reinforcement Learning 361
Automating Sequential Decision Making 362
Summary 363
Glossary 365
Bibliography 367
Index 377
[/spoiler]