Open book draft
Modern Financial Forecasting and Decision Systems
From time-series models to automated agentic systems. Forecasting, optimal decisions, dynamic models, and AI agents in one reproducible workflow.
Start readingContents
- Chapter 00Preface
Preface
- Chapter 01Introduction
Forecasting · Optimal Decision · Modelling Dynamics · AI Agent · Synthetic Data
- Chapter 02Theoretical Basics
Returns and Distributions · Forecasting and Evaluation Basics · Portfolio Theory · Classical Time-Series Models · Estimation and Simulation
- Chapter 03Python Foundations for Financial AI
DataFrames with Polars · Financial Data Visualization · Matrices and Tensors with NumPy and PyTorch · Time-Series Preprocessing with sktime · Time-Series Forecasting with sktime
- Chapter 04Forecasting
Univariate Forecasting · Multivariate Forecasting · Tree-Based Forecasting · Deep-Learning-Integrated Forecasting · Transformer-Based Forecasting · Probabilistic Forecasting · Hierarchical and Cross-Sectional Forecasting · Foundation Forecasters in Practice
- Chapter 05Optimal Decision
What Is Optimal Decision? · Economic Optimal Decision · Dynamic Programming · Optimal Policy · Deep Q-Network · Risk-Aware Reinforcement Learning · Execution and Market Making · Multi-Asset Portfolio RL Benchmark
- Chapter 06Modelling Dynamics
What Is State Space? · Finding Latent State · Dynamic Factor Model: Practical Guide · Factors and Autoencoders · Deep Dynamic Factor Model: Practical Tutorial
- Chapter 07LLM Integration
Why LLM? · Reprogramming Approach · Fine-tuning Approach · Foundation Models · Prompt Engineering for Finance
- Chapter 08AI Agent
LangChain and LangGraph · Tool Calling and MCP · Multi-Agent Topologies · Production Deployment · Evaluation and Benchmarks
- Chapter 09RL Fine-Tuning
Why RL Fine-Tuning · The TRL Library · Reward Modelling · SFT to DPO Pipeline · GRPO and Reasoning
- Chapter 10Synthetic Data
Diffusion · GAN-Based Generators · VAE Generators · Bootstrap and Copula Synthesis · Evaluation and Privacy