So when I got close to the ending for the manuscript I just completed, I decided to do something different: I didn't write it. I knew I was going to get stuck writing the conclusion, and that whatever I came up with would suck, so I got 2/3 through the story, wrote a terrible ending that was way too fast-paced for the story, and started revising.
As I revised, I was better able to understand my plot and wrap my mind around which of the three different endings I'd thought of would work best with the story. I planned more conflicts for the second and third acts as I rewrote the early parts, and by the time I wrote the ending for real, I was pleased with it.
I realize that not everyone sucks at endings like I do, but I think everyone has their own Achilles' heel. Even the best writers have the part that they're least-awesome at. Maybe it's witty dialogue. You know you need some, but when the moment comes to write it, it always falls short. Maybe it's scenery—you see it in your head, it just never makes it onto paper. Maybe it's sentence rhythm. No matter what you do, your sentences fall flat when read aloud.
My advice: skip it. Even if it were beginnings or middles that I struggled with, I still would have skipped my weakest part (with only a vague sketch to get out my worst ideas) as I wrote my fast draft. If you really need witty dialogue, but slowing down to think of something will take you a few days, just write "[insert witty dialogue here--Karen zings Horace]" and come back to it later. If you can't picture a scene, but you need your reader to, write "[insert fitting description of room]" and then keep going. That's the important part: understand why you're stuck, make a note to fix it later, and keep going.
In my most recent book, I had so many brackets--research I needed to do, names I couldn't remember offhand, dialogue I couldn't get right the first time... and missing scenes after the midpoint. When I did my first revision, I could just search for opening brackets and look at what I needed to fix. With my editing brain on, I was able to come up with a solution that I wouldn't have managed during the original drafting.
When I remember that not everything has to go down perfectly in the first draft, and give myself permission to save research and "hard stuff" for later, then the most important part of the process happens: I get the book written.