> For the complete documentation index, see [llms.txt](https://asteria.gitbook.io/asteria/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://asteria.gitbook.io/asteria/protocol-layer/realized-volatility-prediction.md).
# Realized Volatility Prediction
The volatility parameter σ in the option pricing formula is the core parameter in the Asteria quotation and hedging system. The capability of accurate realized volatility prediction plays a key role on whether market maker's risk can be completely hedged or even profitable.
Asteria’s RV prediction model:
Assume the valid option time span 
，system would need time series mode to predict the value of 
• For the real market data，system needs at 
, relatively accurately predict .
• The estimation of price of underlying asset between time 
:
Where ， 
as sampling period for calculating volatility，
When doing back testing, we use the current 
to predict next period .
Asteria will implement the following four volatility time series prediction models, and dynamically adjust the selection of models according to the back testing result, or weighted to obtain the final volatility prediction:
• Directly use historical volatility prediction 
, where the HV can take different sampling periods or time windows of different lengths.
• Use index-weighted historical volatility prediction, 
• GARCH Model
• Stochastic Volatility Model