Название: Applied Linear Regression for Business Analytics with R: A Practical Guide to Data Science with Case Studies
Автор: Daniel P. McGibney
Издательство: Springer
Год: 2023
Страниц: 286
Язык: английский
Формат: pdf
Размер: 10.1 MB

Applied Linear Regression for Business Analytics with R introduces regression analysis to business students using the R programming language with a focus on illustrating and solving real-time, topical problems. Specifically, this book presents modern and relevant case studies from the business world, along with clear and concise explanations of the theory, intuition, hands-on examples, and the coding required to employ regression modeling. Each chapter includes the mathematical formulation and details of regression analysis and provides in-depth practical analysis using the R programming language.

In modern statistics, the fundamental concepts of linear regression have changed little over the past decades. More work has been done to develop and expand on these existing concepts, and regression continues to be an active area of research. Today, many linear regression methods are broadly considered Machine Learning or, even more broadly, Artificial Intelligence. Traditional linear regression analysis can be used to make predictions and inferential understanding. Machine Learning aims to make predictions, sometimes using a “black-box” approach and typically not used for inference. Despite this pitfall, Machine Learning is gaining popularity because predictions are needed with the copious amounts of data being collected in this modern age. Data scientists are required to have statistical, programming, and domain knowledge. To become a data scientist or develop expertise in business analytics, the reader must learn many powerful tools to understand and manipulate data. Linear regression and the R programming language are a few of these powerful tools.

Business analytics and data science problems are solved using various software, most notably R and Python. R and Python are relatively new compared to older programming languages such as C, C++, and Fortran. They also both offer a great deal of flexibility in coding data. One key advantage of R over Python is the number of functions available in “base R.” For example, to load a data file into R, one can use the read.csv command, but in Python, it is necessary to use the Pandas library. These functions from base R make a difference in coding with relative ease. For Deep Learning and Natural Language Processing (NLP), Python is a popular choice. However, for linear regression analysis, we recommend R.

R was created for statistical computing, making it a natural choice to do regression, logistic regression, time series, machine learning, and general statistical analysis. In the case of linear regression, base R is sufficient to perform the analysis and is quite powerful in doing so. In this book, base R is used more frequently than functions from R packages, but a select few packages are used in a few cases. Other key advantages include R’s usage of Posit (formerly RStudio), and the reporting tools available in R. R provides some great tools for the analyst to get going on analyzing data sets. While Python offers many advantages as well, for regression analysis, we recommend using R over Python. It is highly recommended to the reader that after studying this text and programming examples, they continue their studies in both R and Python, as both have advantages and limitations.