Comment by sobriquet9

Why estimate PDF through histogram then convert to CDF, when one can estimate CDF directly? Doing so also avoids having to choose bin width that can have substantial impact.