Machine learning portfolio simulating and optimizing apparatuses, methods and systems

A machine learning portfolio generating apparatus, comprising:

• at least one memory;
• a component collection stored in the at least one memory;
• any of at least one processor disposed in communication with the at least one memory the any of at least one processor executing processor-executable instructions from the component collection, storage of the component collection structured with processor-executable instructions comprising:
• generate a set of simulated market scenarios datastructures with a variational autoencoder via a network accessible server,
• in which the variational autoencoder datastructure is structured as including set of latent variables generated via a neural network encoder, and
• in which the set of latent variables are simulated with neural networks as decoder such that decoding of the simulated market scenarios datastructures follow dynamic dependencies and volatilities of historical market risk factors;
• distribute distribution and dependency joint distribution structures with a transfer layer between the encoder and the decoder,
• in which the latent space variables are structured adopting the distribution and dependency joint distribution structures;
• tuning a transfer layers codec,
• in which the transfer layers codec is the encoder and the decoder and the transfer layer in between,
• in which the tuning is structured including:
• a number of the latent space variables, a number of neurons in the transfer layers codec, a number of lovers of the transfer lavers codec,
• in which the tuning is structure as optimizing an overall fit as between the set of simulated marker scenarios datastructures and a set of historical market scenarios,
• quantify unprecedentedness as a fitted polynomial degree 2 curve, via at least one processor on cloud computing infrastructure, which captures the relationship between movements in VIX and the number of risk factors that experienced unprecedented magnitude of changes;
• and
• train the conditional dependency structure of large movements in VIX,
• in which the VIX up and VIX down levels are solved by the objective function of minimizing the mean squared error between a simulated polynomial degree 2 curve and a historical polynomial degree 2 curve,
• in which the fitted polynomial degree 2 curve is fitted from simulated market scenarios datastructures generated with at least one of:
• a variational autoencoder deep learning model, a guassian copula conditional on large VIX movements.

A machine learning portfolio generating apparatus, comprising;

• at least one memory;
• a component collection stored in the at least one memory;
• any of at least one processor disposed in communication with the at least one memory, the any of at least one processor executing processor-executable instructions from the component collection, storage of the component collection structured with processor-executable instructions comprising:
• generate a set of simulated market scenarios datastructures with a variational autoencoder via a network accessible server,
• in which the variational autoencoder datastructure is structured as including set of latent variables generated via a neural network encoder, and
• in which the set of latent variables are simulated with neural networks as decoder such that decoding of the simulated market scenarios datastructures follow dynamic dependencies and volatilities of historical market risk factors:
• distribute distribution and dependency joint distribution structures with a transfer layer between the encoder and the decoder,
• in which the latent space variables are structured adopting the distribution and dependency joint distribution structures;
• tuning a transfer layers codec,
• in which the transfer layers codec is the encoder and the decoder and the transfer layer in between,
• in which the tuning is structured including:
• a number of the latent space variables, a number of neurons in the transfer layers codec, a number of lavers of the transfer layers codec,
• in which the tuning is structure as optimizing an overall fit as between the set of simulated market scenarios datastructures and a set of historical marker scenarios,
• calculate a set of expected returns for a set of securities, each expected return in the set of expected returns structured as calculated for a security during a simulated market scenario in the set of simulated market scenarios datastructures with:
• the respective security's conditional Beta during the respective simulated market scenario, determined with a set of decision tree ensembles, trained estimating conditional Beta of the respective security, based on a first subset of simulated market factors, and
• the respective security's conditional default probability during the respective simulated market scenario, determined with a set of decision tree ensembles, trained estimating conditional default probability of the respective security, based on a second subset of simulated market factor values.