**In a heated argument, the regression shouted at correlation, “You are dumbass, you don’t even know the cause and effect of this argument”**

**Correlation**is used to measure strength of the relationship between two variables. It can be positive, negative or zero.

**Positive Correlation :**Both variables tend to move in the same direction: If one variable increases, the other tends to increase. If one decreases, the other tends to decrease.

**Negative Correlation :**Both variables tend to move in the opposite directions: If one variable increases, the other tends to decrease, and vice-versa.

*Note: The correlation coefficient may take on any value between +1 and -1.*

**Examples**

1: How are sales of ABC Company and GDP related?

2: How are annual returns on Treasury Bills and Bonds related?

*Suppose you would like to know whether there is a relationship between grades and number of hours you spend studying.*

*You can*

**download the workbook**used in this example and practice what you learn.

**Using Excel:**

**1. If you haven’t already installed the**

**Analysis ToolPak**, Click the

**Microsoft Office**button, then click on the

**Excel Options**, and then select

**Add-Ins**, Click

**Go**, check the

**Analysis ToolPak**box, and click

**Ok**.

**How to Install Analysis ToolPak**

2. Select

**Data**tab, then click on the

**Data Analysis**option, then selects

**Correlation**from the list and Click

**Ok**.

**[Data tab >> Data Analysis >> Correlation]***3. Select the*

**data range**(both independent and dependent variable) in the

**Input Range**box.

4. Check

**Labels in first row**and enter range in the

**Output Range**box and Click on

**Ok**.

**Interpretation:**

As you see the correlation between the grades and number of hours you spend studying is a very positive correlation (84%). This means as more number of hours students study their grades improve.

You can accomplish the same task using

For the above problem the syntax would be

You can accomplish the same task using

**CORREL**function.**Syntax:****=CORREL (range1, range2)**For the above problem the syntax would be

*=CORREL(OFFSET($A$2:$A$500,,ROWS($1:1)-1),OFFSET($A$2:$A$500,,COLUMNS($A:A)-1))**For a detailed explanation of this formula , visit this link***Formula Explained:Correlation Matrix****Do You Know?**

**How to deal with outliers when doing correlation?**Solution: Look at the residuals from a regression by plotting the points to a scatter diagram. If they are not normally distributed around 0 the realibility of the Pearson correlation could be unreliable.

Post a comment