# Empirical Rule (68-65-99 rule) Who to Use in Your Data Set

The empirical rule in mathematics is that in a normal data set, every virtual piece of data will be within standard deviations that are in three deviations of the mean, an average of all the numbers is called mean, this was created by Sal Khan… Rule 68-95-99 is another name for **the Empirical Rule**, it is also called the Three Sigma Rule. The reason why this rule has 68 – 95 – 99 because the three standard deviations are divided in the following manner through the percentage of data sets-

- 68% of complete data rest – The first deviation
- 95% of complete data in two of the standard deviations – Second deviation
- 99.7% of complete data in three of the standard deviations (which means the rest of 0.03% data will be used for accounting of outliers and this is present in every dataset) – Third deviation

The meaning of this is simple, the center of the dataset should contain mean, mode and median.

**How can you determine the Standard Deviation?**

This rule called empirical is specially used for only the prediction of the outcome by understanding a data set. For the prediction or use of the formula, you must calculate the standard deviation by using this formula in these mentioned steps.

- The data sets should be determined, that is the data sets total should be divided by numbers of quantity.
- The numbers in the set that is each number in the set should subtract the mean and then resulting numbers should be squared.
- Determine the mean of each number by using the values that are squared.
- The means that are determined above you should find the square root for the same.

Just exclude the percentage of the minor outlier’s rest; all the majority of the data from the set should fall in the standard deviation that is between the primary percentages of the distribution.

**How can you rule the Empirical Rule?**

As you have read above the empirical rule is purely used to predict the outcomes from the given data set. You have to just determine the standard data’s deviation once it is done the data can be used in empirical rule by the distribution of the data.

Prediction is possible through this formula only because you can assume where the data will fall within the set which you do on the bases of the 68-95-97 rule.

You must use the **empirical rule** where you want to predict the outcome where complete data is not available. This simply helps a lot for those who are studying the data and assume where the data will fall so that it is easier when you get all of the data, this is especially for these types of statisticians.

If the data does not fit in the empirical rule then it is known that the data set is not normal, hence you know that the distribution is not normal and then you must calculate everything accordingly.