A portfolio comprising the ETFs of the market indices of USA, Hong Kong, Singapore, Malaysia, Indonesia, India and Thailand optimized with SLSQP (Sequential Least Squares Quadratic Programming.)
Data as of 31 Dec 2024

Introduction
We anxiously await the 20 January coming of the unpredictable MAGA man and hold our breathe to see if he roils global financial markets with his policies and actions. To cope with this unpredictability, perhaps it might be wise to construct a low-risk portfolio of country market index ETFs.
Let’s do this step by step. First, lets look at the Correlation Matrix of each country’s stock with the Dow Jones Industrial Average. The Covariance Matrix using 250 days’ time series of the Close of the DJIA and the 6-country market index was converted into a Correlation Matrix and here we present it as a Heatmap where:
1.0 indicates perfect positive correlation.
-1.0 indicates perfect negative correlation.
0.0 indicates no correlation
Â

Here we can see that Singapore and Hong Kong have the highest correlation obviously because of their highly open economies and small populations with little room for domestic consumption to contribute to GDP. Indonesia and Thailand have the lowest correlation with the DJIA.
Methodology for creation of low-risk portfolio of ETFs.
We will create an ultra-low-risk portfolio by using ETFs and by maximizing the Sharpe Ratio. But for the risk-free rate in the Sharpe Ratio formula, I think with the utter unpredictability it’s more accurate to be 0%. Not the Fed Funds Rate of 4.33%.
Use the correlation matrix to identify the interplay between the indices and the DJIA.
Allocate weights to balance low correlation assets (e.g., JKSE, KLSE) with higher correlation ones for stability and returns.
Optimize by using the SLSQPÂ algorithm to ensure weights summed to 1 and remained between 0 and 1. See Addendum on SLSQP (Sequential Least Squares Programming) at the bottom of this post. The resulting portfolio will minimize risk while maintaining a reasonable return, evidenced by a strong Sharpe Ratio.
Here are the optimized weights representing the % (in dollar terms ) allocation in the portfolio:

Simulated performance of portfolio over 252 trading days using Monte Carlo Simulation of 1000 Runs

Annualized Return: 8.43%
Annualized Volatility: 4.99%
Sharpe Ratio (Risk-Free Rate = 0): 1.69
Insights:
The portfolio minimizes volatility while maintaining a stable return.
Low-correlation indices help offset risk from highly correlated markets.
Recommendations:
Use this low-risk portfolio approach to navigate uncertain financial environments.
Monitor global events and rebalance the portfolio as needed.
ADDENDUM: The SLSQP Algorithm
The SLSQP (Sequential Least Squares Programming) algorithm is an optimization method commonly used for solving constrained nonlinear optimization problems. Here's a breakdown of what it is and how it works:
1. Key Features
Type: Gradient-based optimization algorithm.
Purpose: Solves problems where the objective function and constraints are nonlinear.
Constraints: Can handle both equality and inequality constraints, which makes it versatile for portfolio optimization problems.
2. How It Works
The SLSQP algorithm uses the following steps:
a. Sequential Quadratic Approximation: The algorithm approximates the original nonlinear problem with a series of simpler quadratic subproblems. These subproblems are easier to solve and provide step-by-step solutions toward the global optimum.
b. Constraints Handling:
Ensures that the weights: Sum to 1 (equality constraint). Stay within the bounds (inequality constraint, e.g., 0 ≤ weights ≤ 1).
c. Iterative Improvement: Starts with an initial guess (e.g., equal weights) and iteratively refines the solution by solving the quadratic approximations.
Continues until the objective function converges to its minimum value or meets a defined tolerance.
3. Why Use SLSQP for Portfolio Optimization?
Constraint Flexibility: It can enforce the constraints on weights, such as ensuring they sum to 1 or remain non-negative.
Efficiency: Handles large-scale problems with nonlinear objectives and constraints effectively.
Gradient Utilization: Uses gradients of the objective function and constraints to find solutions quickly and accurately.
Applications in Portfolio Optimization. In this context, SLSQP is used to: Minimize portfolio volatility (or maximize the Sharpe Ratio) by adjusting weights. Ensure the portfolio weights sum to 1 and stay within the range of 0 to 1.
Comments