<div class="exam-nav"
data-color="#e11d48"
data-prev="5|Exam 5|Exam 5 (CAS).md"
data-current="PCPA|Exam PCPA (CAS)"
data-next="6|Regulation and Financial Reporting|Exam 6 (CAS).md"
data-tracks="ACAS|Associate of the Casualty Actuarial Society (ACAS).md"
</div>
## Exam PCPA Syllabus
<div class="download-dropdown" data-color="#2563eb" data-label="Download Resources" data-files="2025 Exam PCPA Content Outline|https://www.casact.org/sites/default/files/2024-05/Exam_PCPA_2025_F_Content_Outlines.pdf, PCPA FAQ|https://www.casact.org/sites/default/files/2024-05/CAS_PCPA_FAQ.pdf"> </div>
- [Register for Exam PCPA](https://www.casact.org/exams-admissions)
The **Property and Casualty Predictive Analytics (PCPA)** assessment has two components: a 2 hour computer-based exam with 40 questions, and an offline predictive analytics project. Together they measure a candidate's foundational expertise in predictive analytics for P&C insurance. Some prerequisite knowledge is expected:
- Mastery of the concepts in [[Exam MAS-I]], [[Exam MAS-II]], and [[Exam 5]].
- Proficiency in a statistical programming language such as R, Python, or SAS.
- Knowledge of [[Generalized Linear Models]] and predictive modeling techniques.
### Learning Objectives
> [!example]- A. Dealing with Data {Exam 35% · Project 30%}
>
> ### Learning Objectives
>
> Evaluate the dataset and manipulate it so it can be used in a [[Predictive Analytics|predictive analytics]] model.
>
> 1. Gather and assess the relevance of information from stakeholders in actuarial analysis.
> 2. Import, manipulate, and evaluate datasets using generally available programming languages and software packages (e.g., .csv file).
> 3. Evaluate the need for [[Variable Transformation|variable transformation]] and apply appropriate transformations to data.
> 4. Identify and appropriately manage [[Outliers|outliers]] and [[Missing Data|missing data]].
> [!example]- B. Model Diagnostics & Selection {Exam 35% · Project 30%}
>
> ### Learning Objectives
>
> Create and refine a [[Generalized Linear Models|GLM]] model.
>
> 1. Create and run a [[Generalized Linear Models|GLM]] model.
> 2. Evaluate and improve a [[Generalized Linear Models|GLM]] model (e.g., create and interpret diagnostics, conduct [[Cross-Validation|cross-validation]], incorporate [[Offsets|offsets]], mitigate [[Multicollinearity|multicollinearity]] issues, avoid [[Underfitting|underfitting]] and [[Overfitting|overfitting]]) for the data provided and business goals.
> [!example]- C. Model Interpretation & Presentation {Exam 30% · Project 40%}
>
> ### Learning Objectives
>
> Interpret the findings from a [[Predictive Analytics|predictive analytics]] model and present findings to technical and non-technical audiences.
>
> 1. Create and interpret statistical/tabular and graphical/visual representations of data.
> 2. Communicate project technical information, including details on methodologies, modeling decisions, and interpretation of output.
> 3. Communicate project findings to non-technical audiences, including the implications on business outcomes or decisions.
### Project Requirements
Candidates who pass the PCPA exam are eligible for the project component. In each project window, candidates receive a business problem and dataset(s) and must:
- Conduct [[Exploratory Data Analysis|exploratory data analysis]], identify and manage data issues (e.g., transformations, [[Outliers|outliers]], [[Missing Data|missing data]]), and select appropriate target and predictor variables.
- Build a [[Generalized Linear Models|GLM]] (e.g., Gamma, Poisson, Binomial, Log-Normal, Tweedie) to address the business problem, evaluate performance, and iteratively improve the model.
- Interpret findings from both a technical perspective and for the business decision, and submit a technical report (max 1,000 words), code (R, Python, or SAS), and up to 5 supporting tables/graphics.
### Sources
|Source|Abbreviation|Coverage|
|---|---|---|
|[[ASOP No. 23 - Data Quality (2016)]]|ASOP 23|Domain A|
|[[ASOP No. 56 - Modeling (2019)]]|ASOP 56|Domain A|
|[[How Charts Lie (Cairo - 2020)]]|Cairo|Domain C|
|[[Generalized Linear Models for Insurance Rating (Goldburd - 2020)]]|GLM (Monograph)|Domain B — Chapter 7|
|[[Generalized Linear Models for Insurance Data (De Jong - 2008)]]|De Jong and Heller|Domain B — Chapters 5, 6, 8 and related code in Appendix|
|[[R for Actuaries and Data Scientists (Fannin - 2020)]]|Fannin|Domain B — Chapters 6, 15, 18, 19|
|[[Show Me the Numbers (Few - 2012)]]|Few|Domain C|
|[[Predictive Modeling Applications in Actuarial Science Vol 1 (Frees - 2014)]]|Frees, Derrig and Meyers|Domain B — Chapter 6|
|[[Infovis and Statistical Graphics (Gelman - 2012)]]|Gelman and Unwin|Domain A|
|[[R for Data Science (Grolemund - 2017)]]|Grolemund and Wickham|Domain A|
|[[Storytelling with Data (Knaflic - 2015)]]|Knaflic|Domain C|
|[[Python for Data Analysis (McKinney - 2022)]]|McKinney|Domain A|
|[[Modern Applied Statistics with S (Venables - 2002)]]|Venables and Ripley|Domain B — pp. 172–176|