UTRADE Robo seeks to deliver an automated fund management service that maximizes investment return for each client's risk profile, which is determined by their financial objective, as well as risk tolerance.
- The 6 steps of our investment methodology are as follows:
- Identify the set of major asset classes which are investable
- Select low cost ETFs to represent each asset class
- Use Modern Portfolio Theory to allocate among the chosen asset classes to determine the maximum expected return for a given level of risk
- Establish the client's goals and risk profile to assign a suitable portfolio
- Implement and periodically rebalance the portfolio
- Recalibrate the risk level for the portfolio assigned to the client as the client progresses through their goals or if the client provides updates to their goals or risk profile
Asset Class Selection
We have identified Equities, Fixed Income and Commodities as the broad asset classes from which portfolios will be constructed. Within each broad asset class, we select representations from a variety of geographies and risk segments which we assess to be investible and sufficiently diversified.
Fund Selection
Our strategy for choosing the right fund to represent an asset class is to choose the lowest cost fund based on the sum of their costs, including fund expenses, and expenses incurred by the investor.
The criteria for selecting ETFs include:
- Representative of asset class
- Low Expense Ratios
- Tax Efficient
- Market cap weighted, no sector tilts
- Physical Replication
- Liquid — Net Asset Value cut off of US$1b (except for Singapore-focused Funds)
In selecting the funds for our portfolio, our first step is ensuring that the funds we screen for are representative of their asset classes, covering a significant part of the specified asset class. For example, for US equities, we have selected a fund which tracks the S&P 500 index, which covers 80% of the total market capitalisation for US stocks.
Mean-variance optimization using Modern Portfolio theory
Mean-variance optimization (MVO) to allocate assets in your portfolio. MVO is based on an important and influential financial concept called Modern Portfolio Theory (MPT), put forth by Harry Markowitz, who won the Nobel Prize in Economics for his work on MPT. MPT is a theory on how investors can reduce risk for the same portfolio return through diversification.
In summary, the theory argues that for a given level of expected return (the "mean" return), a portfolio's risk or "variance" of returns may be reduced by mixing assets which are not correlated.
There have been criticisms of the theory, such as when correlations rose during the 2008 financial crisis. But we believe that the key lies in the implementation, and the method remains relevant and suitable for investors today.
Capital Market Assumptions
Mean-variance optimization (MVO) requires, as inputs, estimates of standard deviation, correlation and expected return for each asset class.
There are several options for portfolio managers to forecast expected returns for inputs into:
One way is to use historical asset class returns as the estimate for expected returns. However, there are some pitfalls to using historical returns as an estimator. Firstly, past performance is not indicative of future results, and this method risks over-weighting asset classes which have done well previously but are currently overvalued. On top of impacting the portfolio's asset allocation, investors may be misled into expecting that their future returns will match historical performances. For example, in today's very low interest rate environment, it is unlikely that bonds will generate returns equal to what they have done historically. Separately, expected return assumptions are also critical to projecting whether the investor will be able to meet their goals, especially over the long run — and using historical assumptions would likely not be prudent.
Our implementation of MVO incorporates fundamental and quantitative analysis to forecast expected returns for various asset classes. Our portfolio management team will take a forward looking approach, taking into account fees, valuations, interest rates, currency risk premiums, yields and expected growth as part of our of estimate of expected returns.
We utilize historical volatility and correlation data for each asset class to estimate expected volatility and correlation.
Establishing the investor's goal and risk profile
Our profiling engine engages the investor to determine his financial objectives and risk tolerance, and assigns a suitable portfolio risk level.
The time horizon of the investor is a key determining factor of risk. In general, a longer time horizon would result in a higher risk level assigned. Our algorithm will assign the lowest risk portfolio for an investor with a time horizon of two years while the highest risk portfolio will be assigned to an investor who is saving for a goal in 30 years or more.
Expected future contributions and withdrawals to the investment account are also taken into account. Consequently, higher expected future contributions and longer withdrawal paths results in higher risk levels assigned to the investor.
Implementation and Rebalancing
Following the purchase of an initial portfolio, the portfolio will be rebalanced when the investment account is topped up, every six months or at the discretion of UOB Kay Hian's portfolio managers. Generally, our portfolio managers will not employ a rebalancing transaction unless market conditions are volatile and we believe that there is a significant benefit in doing so.
Recalibration
As time passes, the investor will progress towards their goals and their risk levels will be adjusted accordingly. We call this process recalibration. The main goal of doing so is to reduce volatility as one reaches their goal so that there is more certainty of the final outcome.