/quant-finance

Open souce quantitative finance models and algorithms with tutorials

Primary LanguageJupyter NotebookMIT LicenseMIT

Quantitative finance models and algorithms

v0.3

Collection of models with optimization algorithms for Time series analysis, algorithmic forecasting, quantitative research and risk-management.

Assets pricing and Optimization models

European Option Pricing

  • European option pricing via Monte-Carlo simulation, Black-Scholes model
  • discretization error estimate
  • sensitivity analysis of option price to strike and volatility
  • sensitivity of discretization error to number of simulations

European option pricing

Linear Asset Pricing

  • linear asset pricing: FX income, capital budgeting, floating-rate notes
  • univariate concave nonlinear optimization of IRR-YTM using Brent method and binary grid search on subintervals
  • available as mixed integer programming problem, ready-to-use on NISQ devices

Time series analysis models

GJR-GARCH

  • Glosten-Jagannathan-Runkle GARCH(p, o, q)
  • unsupervised optimization of parameters
  • captures asymmetric shocks (leverage effect)

Seasonal ARIMA

  • ARIMA(p, d, q)x(P, D, Q, s)
  • unsupervised optimization of AR, MA and Seasonal parameters
  • provides one-step-ahead predictions and out-of-sample forecast

SARIMAX

Holt-Winters model

  • triple exponential smoothing
  • cross-validation via Conjugate gradient, TNC
  • in-sample prediction and extrapolation

Holt-Winter model

Smoothing methods

  • Moving average
  • Exponential smoothing
  • Double exponential smoothing

Smoothing methods

License and Copyright

Copyright (c) 2019 Oleksii Lialka

Licensed under the MIT License.