by Kshama Dwarakanath, Svitlana S Vyetrenko, Tucker Balch
Continue on SelfMadeQuant
Suggested foundations for optimisation and numerical methods and trading research and honest backtests. Specialist prerequisites vary by paper.
Read the linear algebra used to define a numerical problem.
Identify the objective, constraints and conditions for an optimum.
Use only information available at the simulated decision time.
Separate model selection from the final evaluation.
Understand identifiability and fitted model parameters.
Separate model error from unstable numerical computation.
Reconcile signals, executions, positions and P&L.
Choose a reproducible strategy, specify the trading rule, and test what remains after costs and estimation uncertainty.