OptionPricing - Research on OTC options pricing models
=======================================================================================================================
- SVI MODEL =======================================================================================================================
step 1. DATA PREPARATION :
Prepare option data for optimization, using call options/put options/call & put combined by put call parity adjusted rates.
Utilities functions in svi_prepare_vol_data.py
step 2. MODEL CALIBRATION :
Use Quasi-Explicit Optimization (Nelder-Mead Simplex Algorithm) to calibrate model parameters.
Run svi_calibration_params_opt_XXX.py (XXX stands for different dataset in step 1)
step 3. INSAMPLE PERFORMANCE:
Insample pricing error analysis.
Run insample_performance_svi_put.py
step 4. DYNYMIC HEDGE PERFORMANCE:
Dynamic hedge using t-2 calibrated params and t-1 delta to calculate t date hedge error.Hedge could be based on smoothed implied volatility curve (3-day or 5-day) or original ones.
Run hedging_performance_svi_XXX.py (XXX stands for methods and dataset call/put)
=======================================================================================================================