Learn
There are various strategies to maximise returns depending on your specific requirements. You can learn about the different options here. This is just a brief overview, the original papers and additional resources can also be found here if you want to get more into the maths behind these models.
Classic Strategies
These were the original portfolio optimisation strategies inspired by Markowitz's Modern Portfolio theory, which first showed that diversification can increase the risk to reward ratio of a portfolio. The different versions tailor to different preferences and carry with them their own advantages and disadvantages.
Description:
Invests an equal amount into each stock
Pros:
- Simplest strategy to implement
- Portfolio is diversified
- Portfolio is not disproportionately affected by one stock
Cons:
- Ignores the fact that some stocks are more risky than others, such as high growth stocks and controversial stocks tied to influential people
- Does not consider correlations between the stocks, some assets might move together so it results in lower diversification.
Literature:
Description:
Maximises the Sharpe ratio, which is calculated as the excess return / volatility. Excess return is the expected return above the risk-free rate and we use this to select the portfolio with the highest return per unit of risk. The risk-free rate is the return on an investment with no risk, it is usually measured by interest on a high interest savings account or returns on government bonds.
Pros:
- Has the highest risk to return ratio
- Each unit of cash invested in the portfolio gets the best odds
Cons:
- It relies entirely on historical data to calculate expected returns which is highly speculative.
- Affected hugely by outlier stocks which have done well recently but might not be stable investments.
- Often gives outliers a disproportionate share of the portfolio.
- Assumes efficient markets which is unrealistic
Literature:
Description:
Chooses the portfolio with the lowest standard deviation of returns. This should be the most stable portfolio so it should give consistent returns.
Pros:
- Lowest risk of losing money
- Stable predictable returns, or at least more stable than the other portfolios
- Optimal for risk averse investors who prefer stability over growth
Cons:
- Lower returns on average in comparison to other portfolios due to minimised risk
- If one stock has a much lower volatility than the others it could be disproportionately weighted resulting in low diversification
Description:
Ensures each stock holds an equal share of the risk. Essentially, it is more refined version of equal weighted, which accounts for stocks having a different volatility (risk).
Pros:
- More stable portfolio
- More diversified portfolio, not overexposed on any single stock
Cons:
- Usually lower returns than the market, unless using leverage but this increases risk
Literature:
Description:
Invests in stocks with low correlation so that if one goes down the others are less likely to follow
Pros:
- Makes the portfolio independent of any one stock.
- Reduces risk
Cons:
- Asset correlations can and do change over time
Literature:
New Strategies
These are more advanced strategies, while not exactly new, they are more recent developments and are still active areas of research.
Description:
Groups stocks based on correlations and ensures no single group of similar stocks carries too much risk.
Pros:
- Avoids risk concentration in any one group of similar stocks
- Less sensitive to estimation errors than traditional methods
Cons:
- Derives correlations from historical data but correlations change over time
- Very sensitive to outliers which distort the correlations between stocks
Description:
Minimises the biggest loss which occurs in the worst 5% of cases. If the threshold it achieves is -7.6% it means there is a 5% chance the portfolio will lose more than 7.6% in a year.
Pros:
- Minimises losses when the market goes down.
- Possibly easier to understand the risk presented in this form, compared with volatility.
Cons:
- It assumes returns are normally distributed even though it has been proven stock return distributions have fat tails (high kurtosis).
- Extreme events are more common than we think.