When I was designing low noise low drift DC amplifiers, I needed a way to verify and set the frequency break point between the low noise amplifier and the chopper stabilized amplifier used to control 1/f noise and drift for lowest total noise from DC up to a few Hz. An AC measurement would not work because its frequency response would not extend low enough so what I ended up doing was using my best multimeter, something like a Fluke 8505A, in sampling mode measuring the DC value and then calculating the standard deviation of the samples taken over 10 seconds which is conveniently the RMS value of the noise over the input bandwidth. If I ever design a multimeter, I am going to include a function to make this type of measurement.
The frequency break point was adjusted by setting the integrator gain so once I found the value for the lowest total noise, I went back and calculated the frequency break point and compared it to the low frequency noise curves for the separate amplifiers and found it to match exactly what was needed.
That low noise amplifier ended up working so well that it could be used to measure low values of resistance by thermal noise alone.