Error bars in charts you create can help you see margins of error and standard deviations at a glance. They can be shown on all data points or data markers in a data series as a standard error amount, a percentage, or a standard deviation. You can set your own values to display the exact error amounts you want. For example, you can show a 10 percent positive and negative error amount in the results of a scientific experiment like this:
You can use error bars in 2D area, bar, column, line, xy (scatter), and bubble charts. In scatter and bubble charts, you can show error bars for x and y values.

Click anywhere in the chart.

Click the Chart Elements button next to the chart, and then check the Error Bars box.

To change the error amount shown, click the arrow next to Error Bars, and then pick an option:

Pick a predefined error bar option like Standard Error, Percentage or Standard Deviation.

Pick More Options to set your own error bar amounts, and then under Vertical Error Bar or Horizontal Error Bar, choose the options you want. This is also where you can change the direction, end style of the error bars, or create custom error bars.

Note: The direction of the error bars depends on the type of chart you’re using. Scatter charts can show both horizontal and vertical error bars. You can remove either of these error bars by selecting them, and then pressing Delete.
Review equations for calculating error amounts
People often ask how Excel calculates error amounts. Excel uses the following equations to calculate the Standard Error and Standard Deviation amounts that are shown on the chart.
This option 
Uses this equation 
Standard Error 
Where: s = series number i = point number in series s m = number of series for point y in chart n = number of points in each series yis = data value of series s and the ith point ny = total number of data values in all series 
Standard Deviation 
Where: s = series number i = point number in series s m = number of series for point y in chart n = number of points in each series yis = data value of series s and the ith point ny = total number of data values in all series M = arithmetic mean 