What Exam PA actually is (and why practice looks different)
I am Jeffrey Ting, an FSA and CFA, and I built FreeFellow to make solid exam prep free. If you are studying for Exam PA (Predictive Analytics), the first thing to get straight is what the exam is, because it changes what "practice" even means.
Exam PA is not multiple-choice. It is a 3.5-hour open-response exam. You are handed a business data problem, and you type written responses to specified tasks in Microsoft Word, with Microsoft Excel also available. You are graded on the quality and correctness of what you write. This is an Associate-level (ASA) exam, and most candidates take it after SRM.
One more important detail: as of the April 2023 administration, R and RStudio are not available during the exam. All the relevant R code and output are provided in the exam materials. Your job is to read and interpret that output, not to write or run code under time pressure. So the skills that matter are reading model output, reasoning about it, and writing clear, complete answers to what each task asks.
That is why prep for PA cannot look like prep for the multiple-choice actuarial exams. There is no giant answer-bubble bank that mirrors this exam. The practice that transfers is writing responses to real released tasks and then grading yourself honestly against the rubric.
What free PA practice actually exists
Here is what I offer for PA, all free, no signup required just to browse, no credit card.
Graded walkthroughs of every released sitting. I have reproduced each released administration verbatim from the published Project Statements and Model Solutions: October 2023, April 2024, October 2024, April 2025, and October 2025. On top of those five, the just-released April 2026 exam is up too, and its published model solution goes in as soon as it is available. For every task you write your own answer first, then compare it against the model solution, the graders' comments, and a point-by-point rubric. You can find these at the PA walkthroughs.
500 free concept-check questions. These are short items across the syllabus, meant to drill the underlying ideas quickly. They are distinct from the written exam itself. Think of them as a way to confirm you actually understand a concept before you try to write about it under exam conditions.
22 free lessons with AI-narrated audio, plus a formula and technique sheet and a glossary. These give you the conceptual backbone: what each method does, when to use it, how to read the output, and how to talk about it precisely.
All of this lives under the free Exam PA practice page.
How to practice the writing
This is the part candidates skip, and it is the part that decides your score. Reading a model solution and nodding along is not practice. Writing your own answer, cold, is practice.
Here is the loop I recommend:
- Open a task from a released sitting. Read the Project Statement context and the specific ask.
- Type your full written response, the way you would in Word on exam day. Do not peek at the solution. If the task points to R output, work from that output the way you would in the real exam.
- Only then reveal the model solution, the graders' comments, and the rubric.
- Score yourself point by point. Where did you lose marks? Usually it is not because you did not know the method. It is because you did not say the thing the rubric wanted, or you stated a result without the required justification, or you skipped a step the graders explicitly rewarded.
- Rewrite the weak parts until your answer would earn the points.
On grading: free users get a copy-to-AI prompt for each task, so you can paste your written answer and the rubric into your own ChatGPT or Claude and get a structured critique. Fellow-tier users get that grading instantly inside the platform against the same rubric. Either way, the standard you are measuring against comes from the published model solution and graders' comments, not from my opinion.
If you do this across all the released sittings, you will start to recognize the recurring task types: describe a variable and its issues, justify a modeling choice, interpret a coefficient or a tree split, compare models, recommend and defend. The exam reuses these shapes even as the data changes.
How the walkthroughs map to the real exam
Because each walkthrough is reproduced from a real released administration, the mapping is close to one-to-one. The Project Statement is the same business problem candidates actually faced. The tasks are the same tasks. The model solution is the graded answer, and the graders' comments tell you what the graders actually valued and penalized.
What I have added on top is structure: I split each sitting into its component tasks so you can write one at a time, and I turn the model solution and comments into a rubric you can score against. That is the difference between reading a PDF and rehearsing the exam.
A realistic plan: get through the concept checks and a few lessons first so the methods are solid, then spend the bulk of your time writing task responses from the released sittings, oldest to newest. Save one full recent sitting to do under timed, exam-like conditions near the end so you can feel the 3.5-hour pacing.
A note on independence: FreeFellow is an independent project and is not affiliated with the Society of Actuaries. The exam materials I reproduce are the published Project Statements and Model Solutions, and I always encourage you to read the source documents directly as well.
Frequently Asked Questions
Are the Exam PA past exams free?
Yes. On FreeFellow, every released sitting from October 2023 onward is a free graded walkthrough, reproduced from the published Project Statements and Model Solutions. You can browse them without a signup or a credit card.
Where can I find Exam PA sample questions?
The most exam-relevant practice is the released past sittings themselves, which I have set up as task-by-task walkthroughs at /soa-pa. For quick idea drilling, there are also 500 free concept-check questions across the syllabus.
How do I practice the writing on Exam PA?
Open a released task, type your own written response the way you would in Word on exam day, then reveal the model solution, the graders' comments, and a point-by-point rubric and score yourself against it. Repeat across sittings.
How does the grading work?
Each task has a rubric built from the published model solution and graders' comments. Free users get a copy-to-AI prompt to grade their written answer in their own ChatGPT or Claude. Fellow-tier users get instant AI grading against that rubric.
Is there a question bank for Exam PA?
Not a large multiple-choice bank, and that is deliberate. Exam PA is a written open-response exam, so a MCQ bank would not match it. The practice that transfers is writing responses to real released tasks and self-grading.