Automated (and stunning) actuarial reports with R Markdown and ggplot2 by Claudio Torpes Rebelo, Senior Actuary, Risk Management, Swiss Re

In the first-ever R/Insurance webinar, the R Consortium invites three team members of Swiss Re to share their use of R and Upskilling on data handling and communications. Four years ago Swiss Re set out for its actuarial community two ‘capability building’ objectives, to improve actuarial data handling and communication skills. Swiss Re launched a community-based initiative, the Atelier program, helping each other upskill whilst solving real-world business-as-usual problems. In this session, Swiss Re team members shared their approach to upskilling for data analytics before diving into an example of how to create automated (and stunning) actuarial reports.

The initial part of the R/Insurance series:    • Upskilling on Data Handling and Commu...  

#rstats #rprogramming #actuaries #ggplot2 #rmarkdown #insurance

Main Sections

0:00 Introduction
0:17 Automated reports with R Markdown
1:50 What is R Markdown?
2:14 Reproducibility and automation
4:00 Quarterly report: Messy workflow
6:14 Rethink what we call an automated process
6:43 R Markdown file: Ready to knit?
8:35 What about my Python/SQL/JavaScript skills?
9:00 Remarkable Re (RR) - Study case commercial liability insurer
11:00 What type of reports can be produced with this data set? Example
13:52 Approach to automate comments
15:35 Automatic comment generator, other considerations
24:52 HTML is the richest format for communication that supports a variety of features
25:05 Example of a report in HTML
27:37 Can we incorporate Shiny in an R Markdown HTML document?
28:00 Mark meets Shiny
29:40 Parameterized reports, the road to full automation
31:02 Can we automate workflow even further?
31:39 What else? RStudio Connect
32:32 Faster & Better

More Resources

Main Site: https://www.r-consortium.org/
News: https://www.r-consortium.org/news
Blog: https://www.r-consortium.org/news/blog
Join: https://www.r-consortium.org/about/join
Twitter:   / rconsortium  
LinkedIn:   / r-consortium