At the heart of the Pandas library is the data frame, which makes using the Pandas framework interoperable from a skills-building standpoint. Not only will learning the methods in Pandas be valuable within Python, but you can quickly transfer your knowledge of the framework to R or even Spark (for big data applications). Further, the framework itself implemented in Python is beneficial for data analysis, providing numerous helper functions on the data frame object, that include aggregation methods, standard statistical calculation methods, and handy join/merge, and subsetting functionality that all data analysts will likely use. On top of that, it is built on top of Numpy for easy transference between those types for more heavy-duty/actual work or even pushing it up to a higher level of abstraction for more data-viz/communications/analysis work.

For doing mathematical operations Analysis purpose

Ready optimized implementations for basic statistical calculations like central tendancy measures and quantiles.