Beyond the Spreadsheet
Conventional financial modelling assumes a “Base Case” world where variables move in linear, predictable patterns. In the South African context—where load shedding, carbon tax legislation, and currency volatility interact chaotically—this approach fails.
We approach risk not as accounting, but as physics. By modeling the "Regime State" of the market (Stable, Stressed, or Crisis), we can accurately price the probability of extreme tail events.
def simulate_market_state(grid_health, volatility_index):
if grid_health < 0.65:
return "CRISIS_REGIME"
elif volatility_index > 25.0:
return "VOLATILE_REGIME"
else:
return "STABLE_REGIME"
Markov Chain Logic
We utilize Markov Chains to model the probability of jumping between regimes. Just because the market is stable today does not mean it will remain so tomorrow. Our engine calculates the transition matrix for every hour of the 20-year project lifecycle.
Input Vectors
The model ingests 20+ years of historical data, including Eskom generation availability, coal prices, localized solar irradiance, and ZAR/USD exchange rate volatility, to build a synthetic future that respects historical correlations.