nach oben

2021 | Buch

Kapitel lesen Erstes Kapitel lesen

Business Analytics

Data Science for Business Problems

verfasst von: Dr. Walter R. Paczkowski

Verlag: Springer International Publishing

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This book focuses on three core knowledge requirements for effective and thorough data analysis for solving business problems. These are a foundational understanding of:

1. statistical, econometric, and machine learning techniques;

2. data handling capabilities;

3. at least one programming language.

Practical in orientation, the volume offers illustrative case studies throughout and examples using Python in the context of Jupyter notebooks. Covered topics include demand measurement and forecasting, predictive modeling, pricing analytics, customer satisfaction assessment, market and advertising research, and new product development and research. This volume will be useful to business data analysts, data scientists, and market research professionals, as well as aspiring practitioners in business data analytics. It can also be used in colleges and universities offering courses and certifications in business data analytics, data science, and market research.

Inhaltsverzeichnis

Frontmatter

Beginning Analytics

Frontmatter

Chapter 1. Introduction to Business Data Analytics: Setting the Stage

Abstract

Spoiler-alert: Business Data Analytics (BDA), the focus of this book, is solely concerned with one task, and one task only: to provide the richest information possible to decision makers.

Walter R. Paczkowski

Chapter 2. Data Sources, Organization, and Structures

Abstract

I stated in Chap. 1 that information is hidden, latent inside data so obviously you need data before you can get any information. But just saying, however, that you need data first is too simplistic and trivial. Where data originate, how you get them, and what you do with them, that is, how you manipulate them, before you begin your work is important to address.

Walter R. Paczkowski

Chapter 3. Basic Data Handling

Abstract

Small, tidy data sets are used to illustrate concepts and techniques in typical textbook treatments of statistics and econometrics. They always have just a few rows or records and a few columns called variables or features. The data are neat, clean, and available, meaning you are never told about the complexities involved in finding the data let alone importing them into a statistical or programming package. In addition, there is only one dataset. The use of several that may have to be merged or joined is not discussed. Consequently, learners must determine from other sources how to handle “messy” and large amounts of data. These are, in fact, typical operations in Business Data Analytics. They require preprocessing before any analysis can begin.

Walter R. Paczkowski

Chapter 4. Data Visualization: The Basics

Abstract

Data visualization issues associated with the graphics used in a presentation, not in the analysis stage of developing the material leading to the presentation, are discussed in many books. I focus on data visualization from a practical analytical point-of-view in this chapter, not their presentation. This does not mean, however, that they cannot be used in a presentation; they certainly can be used. The graphs I describe are meant to aid and enhance the extraction of latent Rich Information from data.

Walter R. Paczkowski

Chapter 5. Advanced Data Handling: Preprocessing Methods

Abstract

A problem faced by those new to Data Science is getting past the only data paradigm they know: textbook data which are always clean and orderly with no or very few issues as I noted at the beginning of Chap. 3. Unfortunately, real world data do not agree with this paradigm. They are, to say the least, messy. They have missing values, are disorganized relative to what you need to do, and are just, well, a mess. Before any meaningful work is done, you have to process or, better yet, preprocess your messy data.

Walter R. Paczkowski

Intermediate Analytics

Frontmatter

Chapter 6. OLS Regression: The Basics

Abstract

A major part of Business Data Analytics focuses on understanding relationships and then using them to predict most likely outcomes of business decisions. An example of the former is a price elasticity used for repricing an existing product or setting the initial price for a new one. In either case, the elasticity is calculated based on the relationship between price and quantity, allowing for other factors such as income levels, seasons of the year, geographic locations, and so forth.

Walter R. Paczkowski

Chapter 7. Time Series Analysis

Abstract

The time dimension of the Data Cube is a major complication you will eventually face in analyzing your business data because time is a part of most business data sets. The data, a time series, could be for each second because of sensor readings, each minute for a production process, daily for accounting recording, monthly for sales and revenue processing and reporting, quarterly for financial reporting to legal and regulatory agencies, or annually for shareholder meetings.

Walter R. Paczkowski

Chapter 8. Statistical Tables

Abstract

Statistical tables supplement scientific data visualization for the analysis of categorical data. This type of data is exemplified by, but not restricted to, survey data. Any categorical data can be analyzed via tables. I will continue to use the bread baking company Case Study data.

Walter R. Paczkowski

Advanced Analytics

Frontmatter

Chapter 9. Advanced Data Handling for Business Data Analytics

Abstract

In this chapter, I will set the stage for analysis beyond what I discussed in the previous chapters. I covered that material at a high level. Specialized books cover them in greater detail; in fact, whole volumes are written on each of those topics. The ones in this chapter are different. They cover advanced data handling topics.

Walter R. Paczkowski

Chapter 10. Advanced OLS for Business Data Analytics

Abstract

I introduced the twin concepts of supervised and unsupervised learning in the previous chapter as a segue into a discussion about advanced manipulations of the Data Cube. The reason for the segue is that Business Data Analytics involves more than just modeling. It also involves classification. Both are methods for learning from your data, the former in a supervised fashion and the latter in an unsupervised fashion.

Walter R. Paczkowski

Chapter 11. Classification with Supervised Learning Methods

Abstract

I covered modeling as a way to predict either future events (i.e., forecasting) or the outcome of decisions (i.e., predicting) in the previous chapters. Recall that “prediction” is a broad label that includes forecasting as a subset: all forecasts are predictions but not all predictions are forecasts. Predicting in general is a very important function of Business Data Analytics which is why I spent so much time on it.

Walter R. Paczkowski

Chapter 12. Grouping with Unsupervised Learning Methods

Abstract

I will turn my attention to unsupervised learning methods in this chapter. Recall that these methods do not have a target variable that guides them to learn from a set of features. There are still features, but without a target another approach is needed to extract the information buried inside your data.

Walter R. Paczkowski

Backmatter

Titel: Business Analytics
verfasst von: Dr. Walter R. Paczkowski
Verlag: Springer International Publishing
Electronic ISBN: 978-3-030-87023-2
Print ISBN: 978-3-030-87022-5
DOI: https://doi.org/10.1007/978-3-030-87023-2