# Working of 'Interpolation' to impute missing values & the importance of the order of data points for the process

Definition of interpolation - Interpolation predicts values at a point by studying its neighbouring points (within the same column), as opposed to data modeling where all the columns are taken into considering when studying the relationship between points.

Following are the points gathered as part of my research on interpolation & my doubts:

1. Interpolation assumes smoothness & continuity between consecutive data points when connected using curves/lines. So mixing up/sorting of values in the column will affect the interpolation lines/curves & the subsequent results negatively. Hence, we must not modify the order of data points in the dataset.

Doubt - When these data points are plotted on a graph, then the proximity of the values/data points to one another will depend on the magnitude of the values anyways, i.e., values closer in number will lie closer to each other. It doesn't matter how they were ordered in the dataset. So why is there so much emphasis on preserving the original order of the dataset values?

And how practical is it to assume that the person who has prepared the dataset has arranged the values in an order that ensures continuity & smoothness between data points of the column under consideration? We don't know which column has been considered as the key for the ordering of rows.

Note: I have explained my understanding & doubts as detailed as possible. In case my understanding is wrong or my questions aren't clear, do let me know.