Covariance
Learn about covariance, a statistical term used in security and portfolio evaluation
What is Covariance?
Covariance is a statistical term used in security and portfolio evaluation that measures the degree to which two assets move together. A positive covariance means the assets move in the same direction, and the larger the value, the greater one asset moves in relation to the other.
An example of covariance is as follows:
- Stock A has a covariance to Stock B of +1.4
- For every $1 increase of Stock B, Stock A will increase by $1.4
Holding assets with negative or low covariance is preferable for a well-diversified portfolio because it reduces the risk that all assets will move in the same direction at the same time.
Covariance is a statistical measure that shows how two variables move together.
Key Takeaways
- Covariance is a statistical method for determining the relationship between the movements of two random variables.
- Covariance is a tool that investors use to understand the historical relationship between the returns of two equities.
- Investors can use covariance to help choose equities that react to the market in complementary ways, potentially reducing portfolio risk.
- Knowing the differences between covariance, variance, and correlation is crucial in finance and statistics. Covariance and variance measure data fluctuations, while correlation gives a standardized measure of the relationship's strength and direction.
Examples of positive and negative covariance
The following table illustrates the relationships between various economic variables and their impacts:
Variables | Description | Relationship |
---|---|---|
Temperature and Ice Cream Sales | As temperature increases, ice cream sales also increase. People tend to buy more ice cream on hot days. | Positive |
Advertising Spend and Product Sales | When a company spends more on advertising, product sales typically increase due to higher awareness and demand. | Positive |
Unemployment Rate and Consumer Spending | As the unemployment rate rises, consumer spending tends to fall because fewer people have income to spend. | Negative |
Price of Goods and Quantity Demanded | As the price of a product increases, the quantity demanded typically decreases, assuming other factors are constant (law of demand). | Negative |
Formula for Covariance
The covariance formula can be used to determine the direction of the relationship between two variables or stocks:
For a sample data:
cov(X, Y) = Σ((x – x̄) * (y – ȳ)) / (n-1)
For population data, the formula is:
cov(X, Y) = Σ((x – x̄) * (y – ȳ)) / n
Where:
- cov(X, Y) represents the covariance between variables X and Y
- Σ denotes the summation symbol, which indicates that you need to sum up all the values calculated within the brackets
- x and y represent the values of variables X and Y, respectively
- x̄ and ȳ represent the means of variables X and Y, respectively
- n represents the number of observation
Types of Covariance
There are three types of Covariance:
1. Positive Covariance
When two variables tend to move in the same direction, they are said to have positive covariance. In other words, one variable goes up; the other one usually follows suit. Vice versa. Therefore, in this instance, the value of covariance will be positive.
Example
Step 1: List the Data Points
The data points for X and Y are:
X=[1,2,3,4,5]
Y=[2,4,6,8,10]
Step 2: Calculate the Means of X and Y
x̄=(1+2+3+4+5)/5=3
ȳ=(2+4+6+8+10)/5=6
Step 3: Calculate the Deviations from the Mean
Σ(x – x̄)=[1−3,2−3,3−3,4−3,5−3]=[−2,−1,0,1,2]
Σ(y – ȳ)=[2−6,4−6,6−6,8−6,10−6]=[−4,−2,0,2,4]
Step 4: Multiply the Deviations and Sum Them
Multiply each deviation of X by the corresponding deviation of Y:
(−2)×(−4)=8
(−1)×(−2)=2
0×0=0
1×2=2
2×4=8
Sum these products: 20
Step 5: Calculate the Covariance
cov(X, Y) = Σ((x – x̄) * (y – ȳ)) / n
=20/5=4
2. Negative Covariance
Negative covariance is a term we use when two variables tend to move in opposite directions. This implies that if one variable increases, the other will decrease. When this happens, the covariance value will be negative.
Example
X: [1, 2, 3, 4, 5]
Y: [10, 8, 6, 4, 2]
Cov(X,Y)=4−8−2+0−2−8=4−20=−4
3. Zero Covariance
When the two variables don't have a straight-line relationship, their covariance is zero. This means the variables might be related in some way. But it's hard to predict how they'll move together. In this case, the covariance value will be very close to zero.
Example
X: [1, 2, 3, 4, 5]
Y: [7, 1, 9, 3, 5]
Cov(X,Y)=4−4+4+0−2+0=4−2=−0.5≈0
Covariance Vs. Variance Vs. Correlation
Knowing dissimilarities among covariance variance and correlation is essential in finance and statistics.
Covariance and variance are both used to measure the extent to which data points fluctuate from each other. However, they differ in their uses—for instance, variance measures spread along a single axis.
Covariance checks out whether there is any relationship between two variables. When you have different investments in your portfolio, it is important to know how they perform relative to one another.
On the other hand, the correlation coefficient measures how strong or weak a linear relationship exists between two variables, hence, giving more standardized indicators as compared to covariance.
The table below provides a detailed comparison of covariance, variance, and correlation, outlining their definitions, ranges, indications, interpretations, normalization, usage, and examples:
Feature | Covariance | Variance | Correlation |
---|---|---|---|
Definition | Measures the directional relationship between two variables. | Measures the spread of data points along a single axis. | Measures the strength and direction of the relationship between two variables. |
Range | Can be any value (positive or negative) | Always non-negative | Ranges from -1 to +1 |
Indicates | Direction of the linear relationship | Degree of spread around the mean | Strength and direction of the linear relationship |
Interpretation | Positive covariance: variables move in the same direction; negative covariance: variables move in opposite directions. | Higher variance indicates greater spread. | +1: perfect positive correlation; -1: perfect negative correlation; 0: no correlation |
Normalization | Not normalized | Not normalized | Normalized by standard deviations of the variables |
Usage | Financial asset analysis, portfolio diversification | Risk assessment, data dispersion analysis | Statistical analysis, financial market prediction |
Example | Covariance of stock returns to assess portfolio risk. | Variance of test scores to evaluate consistency of performance. | Correlation between height and weight to study relationship strength. |
Free Resources
To continue learning and advancing your career, check out these additional helpful WSO resources:
or Want to Sign up with your social account?