User Pre-Defined Parameterization
User-defined input criteria (Investment Preferences and Risk Tolerances) is “ Parameterized” and passed on to the ATPHIZYOM System AI (Artificial Intelligence model) and utilized by the ATOM-Algorithm or certain techniques within its family of algorithms to process the User instructions, analyze the Global Asset-Universe stochastic data and respond with a corresponding User-inquiry output result.
Asset-Universe Distillation & Ranking by “Numerical Volatility Sequencing” & “Numerical Sharpe-Ratio Sequencing”: Distillation & Ranking of the Global Asset-Universe based on User-defined input criteria (Investment Preferences & Risk Tolerances) to isolate and extract “User-Matching” Assets, via algorithms and methodologies that manipulate a series of Asset data-matrix parameters including both the two key stochastic metrics (a) Stochastic Volatility, AND (b) Stochastic Sharpe-Ratio.
Asset-Allocation by “Numerical Volatility Sequencing”: Distillation-Boundaries synthetically produced by “Brute-Force” Allocation (“Asset-Bucketing”) via the methodology of using either (a) Modular Arithmetic to evenly distribute Assets into Asset-Groups/Asset-Buckets, OR (b) One-Way ANOVA / Paired T-Test (statistical hypothesis testing) of strength and competence of distinct Numerical Volatility Sequences for grouping of Assets by their individual “Stochastic Volatility” ranges/categories.
Monte Carlo Portfolio Weight Simulation
THE OCCUPANCY THEOREM: Asset-Allocation by “Brute-Force” Combinatorics & Permutations is determined by the Binomial Coefficient. Such an application algorithm has a computational complexity or running-time of O(n!) (“Factorial-Time”), where “n” = Number of Allocation-Units [Unit Weight = 100% ÷ n]. However, by limiting the “ Allocation-Units” [or Unit Weight] & “ Asset-Buckets” to a small & fixed-finite number, and using a HEURISTIC APPROACH to obtain “Optimal Portfolio Allocation” along the Efficient Frontier, subject to the User’s MAXIMUM Risk Tolerance (V), artificially forces the application algorithm running-time to demote significantly to a magnitude of O(n) (“Linear-Time”) and thereby significantly reducing the number of Combinations/Iterations. Furthermore, by eliminating the “Trivial” Allocation-Combinations (i.e. focusing on NON-TRIVIAL ALLOCATIONS), the Heuristics Running-Time “Upper Bound” can be further reduced, thus allowing feasibility and capability of the ATOM-algorithm for mass-deployment and scalability.