Thread Subject: Correcting for the phase shift

Hello,
I've got the following problem and I was wondering whether anyone could suggest a solution. I need to analyze data which were recorded using an amplifier which would intorduce a phase shift at the very low frequencies due to the built-in high-pass filter. Unfortunately the phase information at these frequencies is very important for us. However, since the data were recorded some time ago, the amplifier is no longer available. The good news though is that the phase shift was measured: sinusoids of different frequency were injected as an input and the delay was measured. However, other infos are missing such as the length of the impulse response of the filter or the impulse response itself. I was wondering if anyone had any idea of how to effectively correcting for the phase shift relying only on this information and without distorting the data.
Many thanks in advance,
Cesare