Investing in a diversified portfolio of Exchange-Traded Funds (ETFs) has become increasingly popular among investors looking to balance risk and reward. One of the most effective methods for optimizing such portfolios is through Mean-Variance Portfolio Optimization (MVPO). This technique allows investors to construct a portfolio that maximizes expected return for a given level of risk.
The primary aim of this analysis is to utilize quantitative mean-variance optimization to determine the optimal allocation of country-based ETFs. By evaluating individual countries’ ETFs, we seek to provide insights into potential portfolio adjustments, facilitating a strategic allocation away from the All Country World Index (ACWI) on a per-country basis. This approach allows for a more nuanced portfolio construction, potentially enhancing risk-adjusted returns and capitalizing on specific market opportunities.
1. Data Collection and Preparation
For this analysis, historical prices and dividends for a selection of broad country iShares MSCI ETFs were collected using the Yahoo Finance API. The curated list of ETFs includes a diverse set of countries, from major economies like the United States and Japan to emerging markets such as Brazil and Turkey.
The data spans the most recent ten calendar years, with prices recorded at monthly intervals and standardized to USD for consistency and comparability.
2. Visualizing Historical Returns
Historical total returns, inclusive of dividends, were computed for each ETF in the dataset. The 10-year returns, showcased in image below, represent the geometric average total returns over the specified period. Year-to-Date (YTD) returns reflect the total returns accrued within the current calendar year (to 30 April 2024).
Additionally, the frequency distribution of returns was analyzed to identify deciles, which serve as reference points for delineating color gradations on the heat map visualization.
The rest of the article is available on Medium and the powerpoint presentation available to download on Linkedin