I'll wager a WAG and say you'll be closer to an inch than 2. It'll be the arcsin of the distance that needs to be added to the upper arm divided by the distance between the upper and lower balljoints to get you to 0.
This is uncorrect, I screwed this up pretty badly as using 2.5 degrees doesn't give a right triangle in my version so we need to cut it in half. Sin 1.25 = half the distance upper arm needs extended divided by distance between ball joint pivot points. Put another way, half the distance arm needs extended = .0218 divided by the distance between the ball joint pivot points. Since we cut it in half, multiply it by 2 and that's what you need to get to 0.
I like your concept but I find your formula* which includes computing tire diameter causes the process to be less than straightforward. IOW, I physically measure my tire diameter while tire is properly inflated with car level, and on level ground. My current tire is 26" in diameter. If I am trying to get one degree of negative camber then I multiply 26" by 0.01745 (sin of one degree) which equals 0.45 inches**. Then I use a builders square on a level floor to measure difference from vertical (the builders square represents vertical) at the top and bottom of tire sidewall. With negative camber, the top of the tire should be 0.45" closer to the centerline than the bottom of the tire.
*- Tire Diameter ("TD")= Tire Width (mm) * Aspect Ratio/100 *2/25.4 + Wheel Dia.
**- 26inches x sin(1 degree)=26" x 0.01745=0.45"
I agree that using the tire diameter here is not the correct answer as the shims are farther toward the center of the tire so their effect will be magnified, likely close to 2X.