Here’s what you need to know about the pros and cons of bond ETFs.
What is correlation?
Correlation is the growth relationship two securities have between each other. A positive correlation shows that the securities are moving in the same direction and a negative one shows that they’re moving in different directions. This is represented by a value between -1 and 1, with 0 meaning that no correlation exists.
Correlation is expressed by what’s called a correlation coefficient, which is a value between -1 and 1. It can be used to predict an asset growth in relation to another in an investor’s portfolio.
For the sake of diversification, when a portfolio manager picks assets, she may be hoping that there’s a negative correlation between them. That’s means the assets are controlled by discrete market forces, so her portfolio is protected if any one asset should fail. However, it could also mean that growth in one asset is offset in some degree by the other.
The correlation coefficient helps to standardize the measurement of the difference between the performance variables of two separate securities, like stocks or bonds. The value shows how much a security’s growth is dependent on the same variable as the other’s, and the difference between that value out of 1 shows how much that security’s growth is dependent on different factors.
The value of your home may be increasing in relation to your other assets. Is it time to take out a home equity loan?
Becky has a modest portfolio of assets. She wants to better diversify her portfolio by ensuring that no two assets are buoyed by the same market forces. She buys some stock in a company, and calculates its correlation coefficient against stock from an unrelated industry. She finds that the correlation coefficient is -0.2; as the first stock rises, the second stock falls slightly.
Later, Becky compares the first stock to a third stock and finds that the correlation coefficient is 0.4. That tells her that 40% of the stock is affected by the same variables, while 60% is affected by other, possibly unknown variables.