Options Evaluator With an Artificial Intelligence-Based Volatility Model

Abstract

The subject of this paper is an options modeling system, which aims to provide the most accurate profit forecast possible for options portfolios in a comprehensible form, as software on the market will misrepresent this in the absence of accurate implied volatility data, which can put trading success at risk. The software determines future implied volatility from 1-year historical options on VXX stock with 6 samples per trading day using statistics and a deep neural network (Long short-term memory LSTM). Using this statistical approach and the trained volatility model, the system calculates the profit/loss curve, thus providing a more accurate picture of the possible future outcomes of a given portfolio.