On: Pricing Americans with finite-difference

[logicchains]

You can train a small NN on a bunch of synthetic data and the inference will be pretty fast.

[gituliar → logicchains]

Curious to see some references to people doing this. Have any ?

[logicchains → gituliar]

https://www.aimspress.com/aimspress-data/qfe/2023/2/PDF/QFE-... is an example

[gully_boy → logicchains]

was thinking exactly this... and even better, why not on real data, especially since real, clean, data in this field is not at all hard to find.

[gituliar → gully_boy]

Real historical data might not cover current market state.

For example, option price among other params depends on the interest rate. For the last decade interest rate was around 0% in Europe and slightly higher in US. If you train on that data only, there is no chance to "learn" option prices in the high-interest-rate environment which we saw for the last few years. Hence, you need synthetic data to learn that region of the market space.