OPTIMIZATION OF PORTFOLIOS THROUGH MULTITEMPORAL SCREENING AND MOMENTUM STRATEGIES: AN ANALYSIS OF THE SEMICONDUCTOR SECTOR IN THE S&P 500 UNDER SIMULATION VALIDATION

Abstract

This study analyzes the effectiveness of asset selection methodologies (screening) based on multi-temporal returns and cross-sectional momentum within the S&P 500 index. The central problem lies in the manager’s ability to capture alpha in sectors with high technological intensity and idiosyncratic volatility, such as the semiconductor industry. The methodology employed integrates quantitative filters that evaluate time windows ranging from one week to three years, allowing for the distinction between stochastic noise and structural trends. For empirical validation, the results of the simulation identified as “Madison,” executed in the Investopedia Stock Simulator between January and June 2026, are analyzed. The findings reveal an exceptional portfolio performance, with an initial value of 1 million dollars reaching a value of $3,202,633.73, with an effective return of 112.64%. The tactical concentration in assets with persistent momentum such as Micron Technology (MU), Western Digital (WDC), and Applied Materials (AMAT)—with individual returns exceeding 75%—validates the hypothesis that multi-temporal screening optimizes the risk-return profile. However, the impact of the operational leverage used in the simulation and the sustainability of these results under the Adaptive Market Hypothesis are discussed. It is concluded that rigorous screening is a necessary condition for the generation of excess returns in high-beta environments.