An outlier is quite separated from the other data points in a scatter plot, i.e., the distance between the outlier and other data points is significantly larger compared to the difference between any other two data points.
An outlier does not show any essence of the data as it is quite separated from the rest of the data points.
A trend line is a line around which all the given points are equally scattered.
The trend line that captures the essence of the data is called the line of best fit.
∴ We must not consider the outlier for predicting the line of best fit visually.
Consider the following example: An experiment has been conducted to determine the efficiency of a tractor engine by measuring the fuel consumption at the different power outputs of the engine. The data is plotted on a scatter plot as shown below.
We find that there is an outlier present in the data set, marked by a blue arrow. This value of fuel consumption is relatively higher compared to other data points.
Hence, this value is excluded while drawing a line of best fit; otherwise, the line of best fit might get more slanting upward, resulting in the wrong prediction.
This is also true for a statistically calculated line of best fit.
∴ An outlier does not affect the line of best fit.