Bibliographic record
Abstract
Computational Actuarial Science With R by Arthur Charpentier (editor), 2015, Boca Raton, FL: CRC Press, 618 pages, ISBN: 978-1-4665-9259-9. The popularity of R software in data science, statistical analysis, and predictive analytics jobs has grown tremendously in the past decade. Based on the study by Muenchen (2016) on the number of scholarly articles found in 2015 for each commercial software, R software is reported in second place following SPSS, surpassing SAS. A number of books using R have already been written in statistics, economics, engineering, psychology, and other disciplines. Most books written in the actuarial science area focus exclusively on theory while lacking practical applications, especially related to a particular use of computational methods and software. To my knowledge, the first attempt to integrate R with actuarial science applications was made in the book Modern Actuarial Theory With R, written by Kaas et al. (2008), focusing mostly on nonlife insurance topics. The book Computational Actuarial Science With R provides a much broader and comprehensive review of actuarial topics related not only to nonlife insurance but also to life insurance and finance areas of actuarial practice. As the actuarial science field has changed in the past two decades with advances in predictive modeling, modern financial economics, and statistical computing methods, there has been a great need for developing modern actuarial methods that focus on the computational aspects of actuarial science. Implementation of these methods in R software not only allows the actuarial field to keep up with computational science (e.g., computational statistics) but also to remain competitive in the marketplace. Computational Actuarial Science With R elegantly covers a great deal of useful material and applications of R in actuarial science and leaves out much of the actuarial theory that is commonly found in other actuarial books. Numerous real data sets that accompany the book come from 14 countries, bundled up in an R package, CASdatasets, and allow researchers, industry practitioners, and students to get a hands-on, efficient implementation of actuarial concepts and data analysis. A beginning user of R can use the introduction of the book to get up to speed on basic terminology and expressions of the R language. Certain sections of the book can also serve as supplemental material to regular textbooks used with actuarial courses taught at the university level. Computational Actuarial Science With R is divided into four main parts: Introduction to R Language, Statistical Models With R, Methodology, and Life Insurance, Finance, and Non-Life Insurance. As the editor of this scholarly book, Arthur Charpentier, a professor of actuarial science at the University of Quebec at Montreal, has put together a fine collection of articles prepared by 26 contributors (including himself) from the industry and universities around the world. The first section of the book includes several methodological concepts such as statistical inference and learning, Bayesian philosophy, spatial analysis, reinsurance, and extreme events. Drawing heavily on the theory presented in Klugman, Panjer, and Willmot (2012), Chapter 2 presents the most common discrete, continuous, and mixed distributions used in actuarial science and their implementation in R. Here, a reader should be aware when using mixtools and normlmix packages to model mixtures based on normal distributions because they are not suitable for modeling loss data that are typically defined on a positive domain. In the same chapter, the definitions of linear regression model, aggregate loss distribution, copulas, and multivariate distributions are explained, followed by R code illustrating the implementation of these statistical methods. The Bayesian approach to solving actuarial problems with a rich set of R tools is presented in Chapter 3. …
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.057 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.006 | 0.006 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.004 | 0.008 |
| Insufficient payload (model declined to judge) | 0.068 | 0.050 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".