I am trying to calibrate a set of option pricing models (BS, Merton, Kou, Bates) to a set of market option prices using Polygon.io. I want to adhere to the procedure implemented by Persson (2013) (https://lup.lub.lu.se/luur/download?func=downloadFile&recordOId=3809647&fileOId=3809648) so that for each day, I need to extract 20 option prices i.e. 4 maturities and 5 strike prices. The goal is to calibrate the optimal parameter set for each day using a least-squares optimization. I struggle to find daily option prices for say a year so that I can calibrate my models for a sequence of 365 days and see how the optimal set changes throughout this year. Polygon states that it provides 2-y historical data, but I am not able to find the right way of implementing my calibration procedure.
The aggregrate bars and option chain does not provide the desired results, since these are only listed for the most recent date and not historical dates.