We present a new method for modeling derivatives based on a complex-valued extension of Shannon entropy. Instead of classical real probabilities, we use an imaginary entropy that introduces an additional phase dimension. This allows risk to be interpreted not only as uncertainty, but as directed uncertainty. This creates a new strategic fictitious decision axis. The framework is deterministic, interpretable, and programmed entirely in Python. DISCLAIMER (Research Only)This repository contains a research prototype. It is provided for educational and researchpurposes only. It does NOT constitute financial, investment, legal, medical, or any otherprofessional advice. No warranty is given. Use at your own risk. Before using any outputs toinform real-world decisions, obtain advice from qualified professionals and performindependent verification.