Quantum Portfolio Optimisation Using QAOA and QUBO Annealing

Does quantum computing help a portfolio manager today? This project picks K of N real NSE stocks (equal weights, mean-variance objective) by writing the choice as a QUBO and solving it with QAOA on an ideal simulator and on the IBM FakeTorino noise model, against simulated annealing, tabu search and exact enumeration for 8 to 20 qubits, and then back-tests the chosen portfolios month by month against Markowitz, minimum variance and 1/N. The answer, measured and stated plainly, is no: classical tabu search finds the optimum every time in milliseconds, QAOA is barely better than a random guess and worse at 18-20 qubits, and the QUBO portfolios lose to simple 1/N out of sample. Simulators only; MATLAB R2026a code written, Octave-checked; not investment advice.

Category: Quantum Computing - Finance

Quantum Computing Project Q01 from ProTechSC Academy, with graphical abstract, video demonstration and project guidance.