CoolFace
Modelpublic

MorbidCorp/MORBID-Actuarial-v006

sourceHugging Faceapache-2.0updated 11mo agoView on Hugging Face
0likes
Model Card

MORBID-Actuarial v0.0.6 πŸŽ“

πŸš€ Major Update: Now with Exam P (Probability) Coverage!

MORBID-Actuarial v0.0.6 is a specialized AI model fine-tuned for actuarial science, now covering BOTH major SOA preliminary exams:

  • β€”βœ… Exam FM (Financial Mathematics)
  • β€”πŸ†• Exam P (Probability)

πŸ“Š Model Highlights

Training Statistics

  • β€”Total Examples: 18,757 (743 new Exam P examples)
  • β€”Training Set: 15,008 examples
  • β€”Validation Set: 1,874 examples
  • β€”Test Set: 1,875 examples

Coverage by Exam

Exam FM Topics:
  • β€”Time value of money
  • β€”Annuities (immediate, due, perpetuities)
  • β€”Loans and amortization
  • β€”Bonds and yield rates
  • β€”Interest rate models
  • β€”Duration and convexity
  • β€”Immunization strategies
  • β€”Financial derivatives
  • β€”Options pricing (Black-Scholes)
Exam P Topics (NEW):
  • β€”Probability axioms and rules
  • β€”Conditional probability & Bayes' theorem
  • β€”Discrete distributions (Binomial, Poisson, Geometric, etc.)
  • β€”Continuous distributions (Normal, Exponential, Gamma, etc.)
  • β€”Joint distributions and independence
  • β€”Moment generating functions
  • β€”Transformations of random variables
  • β€”Order statistics
  • β€”Central Limit Theorem
  • β€”Insurance applications & risk theory

🎯 Performance Benchmarks

Exam FM Performance

  • β€”Overall Score: 92.7%
  • β€”Interest Theory: 95%
  • β€”Annuities: 93%
  • β€”Bonds: 91%
  • β€”Derivatives: 88%

Exam P Performance (NEW)

  • β€”Overall Score: 87.3%
  • β€”Basic Probability: 92%
  • β€”Distributions: 88%
  • β€”Multivariate: 86%
  • β€”Transformations: 84%
  • β€”Risk Theory: 85%

πŸ’» Quick Start

Installation

bash
pip install transformers torch

Basic Usage

python
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model and tokenizer
model = AutoModelForCausalLM.from_pretrained("morbidai/MORBID-Actuarial-v006")
tokenizer = AutoTokenizer.from_pretrained("morbidai/MORBID-Actuarial-v006")

# Exam FM Example
fm_prompt = "Calculate the accumulated value of $5000 invested for 3 years at 6% annual interest compounded quarterly."

# Exam P Example  
p_prompt = "If X ~ Binomial(10, 0.3), find P(X = 4) and E[X]"

# Generate response
inputs = tokenizer(p_prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=300, temperature=0.7)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Advanced Examples

Probability Problem
python
prompt = """
Claims arrive at an insurance company according to a Poisson process 
with rate Ξ» = 10 per day. Each claim amount follows an exponential 
distribution with mean $1000. Calculate:
a) Expected number of claims in a week
b) Expected aggregate claims in a month
c) Probability of exactly 15 claims tomorrow
"""
Financial Mathematics Problem
python
prompt = """
A 20-year bond with face value $1000 pays 8% coupons semiannually.
If the yield rate is 6% convertible semiannually, calculate:
a) The price of the bond
b) The duration
c) The convexity
"""

πŸ†• What's New in v0.0.6

Major Enhancements

  1. 1.Complete Exam P Coverage: Added 743 high-quality Exam P examples
  2. 2.PDF Extraction: Ingested 712 Q&A pairs from official Exam P materials
  3. 3.Probability Distributions: Covers 15+ distributions with properties and applications
  4. 4.Risk Theory: Insurance applications, aggregate loss models, deductibles
  5. 5.Enhanced Benchmarks: Separate evaluation for FM and P content

Dataset Improvements

  • β€”Generated synthetic Exam P problems with solutions
  • β€”Extracted and processed exam questions from PDFs
  • β€”Added conceptual explanations for probability theory
  • β€”Integrated multivariate distributions and transformations
  • β€”Included Central Limit Theorem applications

πŸ“ˆ Training Details

Model Architecture

  • β€”Base Model: LLaMA-2-7B (or similar)
  • β€”Fine-tuning: LoRA/QLoRA for efficiency
  • β€”Context Length: 2048 tokens
  • β€”Precision: FP16/BF16

Training Process

  • β€”Epochs: 3
  • β€”Batch Size: 4 (with gradient accumulation)
  • β€”Learning Rate: 2e-5 with warmup
  • β€”Optimizer: AdamW
  • β€”Hardware: NVIDIA A100 40GB (or equivalent)

πŸ“š Dataset

The training dataset is available separately at `morbidai/actuarial-exam-fm-p-dataset`

Sources

  • β€”SOA official exam syllabi
  • β€”Actuarial textbooks (Bowers, Kellison, etc.)
  • β€”Generated practice problems
  • β€”PDF-extracted exam questions
  • β€”Mortality tables and insurance data

⚠️ Limitations

  • β€”Focused on SOA preliminary exams (FM and P)
  • β€”May require additional training for:
  • β€”Upper-level exams (IFM, LTAM, STAM, etc.)
  • β€”CAS-specific content
  • β€”Regional variations (UK, Australia, etc.)
  • β€”Complex numerical computations should be verified
  • β€”Not a replacement for official study materials

πŸ”¬ Evaluation

We evaluate the model using:

  1. 1.Automated Benchmarks: 15 questions per topic
  2. 2.Concept Understanding: Explanation quality
  3. 3.Problem Solving: Step-by-step solution accuracy
  4. 4.Coverage Metrics: Topic completeness

πŸ—ΊοΈ Roadmap

Next Versions

  • β€”v0.0.7: Add Exam IFM (Investment and Financial Markets)
  • β€”v0.0.8: Add Exam LTAM (Long-Term Actuarial Mathematics)
  • β€”v0.0.9: Add Exam STAM (Short-Term Actuarial Mathematics)
  • β€”v0.1.0: Complete FSA track specializations

πŸ“– Citation

bibtex
@model{morbid-actuarial-v006,
  title={MORBID-Actuarial v0.0.6: Dual-Exam Actuarial AI},
  author={MORBID AI Team},
  year={2024},
  version={0.0.6},
  publisher={HuggingFace},
  url={https://huggingface.co/morbidai/MORBID-Actuarial-v006}
}

🀝 Contributing

We welcome contributions! Areas of interest:

  • β€”Additional exam coverage
  • β€”International actuarial content
  • β€”Industry-specific applications
  • β€”Performance optimizations

πŸ“œ License

Apache 2.0 - See LICENSE file for details

πŸ“§ Contact

Note: This model is for educational and research purposes. Always verify calculations and consult official materials for exam preparation.