Thanks a lot for your reply!
I am using ni usb 6356 board. Its max analog input sampling freq is 1 M S/sec.
May be you are right. I increased my sampling rate to 900k S/s from 500k S/s. Now, results are better than earlier.
Biggest surprise is this: I replaced capacitor with the resistor = 15K Ohm. Now my circuit looks like voltage divider with equal resistor. Ideally, I should get zero phase difference and -6db magnitude in the bode plot. But I am getting non zero phase which is near to zero degree at lower frequencies and linearly increases to around 15 degrees when frequency of my signal reaches 100Khz !!!
Also, I am getting -6db magnitude for lower frequencies and it increases to -8.5 db as frequency increases towards 100K Hz.
I think you are right in saying that sampling frequency is not enough in my case.
Thanks for the suggestions.