Proper exposure in video has two drastically different answers, based on who you ask.

What M. V says

Vicent Laforet prefer to underexpose his footage, reasoning that it is easier to bring back details from the shadows than from over exposed areas.

What I say

While I agree with Vincent that it is impossible to bring details that have been blown out, my personal approach is to over expose my clip up to the point where clipping would start and back off a bit (using the live histogram). Then, once I am in postproduction, I reduce the exposure to the desired level.

The reason for this is that video compression formats, such as H.264 assigns more memory to store the details in the bright part of an image than the dark parts. So, by shooting a brighter image, the camera will capture and store more details for every frame.

There is one big caveat to this: you have to be extremely careful to not clip your highlights since it is impossible to recover any details from an over exposed section. This is even more challenging when shooting at high ISO since the dynamic range is greatly reduced, making clipping much easier.

While it is hard to manage exposure with much precision using the camera controls, a better way to do it is to use a graduated ND filter while watching the live histogram.

Also, clips shot this way require a more post processing. First you need to correct the exposure, then you have to boost the saturation & contrast to remove the washed out look.

Yes but…

If you are shooting in an uncontrollable/changing environment, you better play it safe and aim for a well balanced histogram. You will not get the best image quality but in these cases, it is better to play it safe than missing the shot.

Follow me!