طراحی سبد بهینه پویای سرمایه گذاری با حداقل ریسک: شواهدی جدید از الگوی خودرگرسیون برداری متغیر در زمان[Dynamic Optimal Portfolio Design with Minimum Risk: New Evidence from the Time Varying Parameter Vector Autoregression Model]

This study aims to design an optimal investment portfolio in order to reduce risk and enhance returns. To this end, daily data of 10 companies listed on the Tehran Stock Exchange over the period from 08/08/2016 to 02/01/2024 were analyzed. The research methodology is based on the Time Varying Parameter Vector Autoregression (TVP VAR) model and employs the Minimum Variance Portfolio (MVP), Minimum Correlation Portfolio (MCP), and Minimum Connectedness Portfolio (MCoP) approaches for portfolio construction. The findings indicate that the portfolio with the minimum variance strategy yields the highest cumulative return. Under normal market conditions, the highest optimal weights are allocated to Damavand (19%), TAPICO (18%), and Seshahed (15%), whereas under bullish and bearish market conditions, the asset composition changes significantly. The analysis of dynamic optimal weights shows that some firms, such as Shetran and TAPICO, play a key role in risk hedging during specific periods. The results emphasize that the use of dynamic and time varying models enhances analytical accuracy and better reflects market realities. Accordingly, training investors in dynamic portfolio management and the use of advanced tools for portfolio adjustment is recommended. The findings have important implications for policymakers and market participants and highlight the necessity of designing flexible investment policies.