Glad you enjoyed the article - did you see the follow up piece that I wrote ? medium.com/analytics-vidhya/how-to-estimate-optimal-stock-portfolio-weights-using-monte-carlo-simulations-modern-portfolio-d27d534e8a1a.

I haven't explored DCC-GARCH models myself but essentially I re used the same approach (either Geometric Brownian Motion or Bootstrapping) and wrote some code that runs simulations multiple times to forecast the future OVERALL portfolio price movement using different weights (which are randomly assigned) and then calculate the P50 realisations for each configuation of portfolio weights ( I guess I could extend the code to plot all the realisations from P01 to P99 rather than only the P50 realisation but it gets very computationally intensive)

Anyway - by doing so , the code can identify the Efficient Frontier of the portfolio and also the (P50?) weights with the "best" risk-return ratio with the highest Sharpe Ratio.

This is available in the Web App https://www.stonksforecast.online under the option Forecast > Portfolio Weights For Optimal Risk-Returns but if you want, I've also made the script available in a Python Notebook here if you want to modify the code itself : https://github.com/ZhijingEu/StockSimulatorFlaskApp/blob/master/MonteCarloEfficientFrontier.ipynb