Do I have to be affiliated with an institution to publish on arXiv? Is arXiv the right place to put it?

Independently of whether you need an affiliation, this sounds like a well-known representation of trig formulas (Euler's formula). There is quite some theory involved with this, including and up to Galois theory. You should check with someone knowledgeable whether you indeed discovered anything new.

It's fun to discover new stuff, and it's a cool feeling to recover things that very smart people came up with, but, once you are there, you have to give it the reality check, which is the ultimate hallmark of the scientist.

Perhaps some friendly mathematician from the local college is happy to spare half an hour with a talented high schooler to give you at least an idea how novel the idea is or what else to try from this point on?

I believe an affiliation is not necessary when registering, although their help page only mentions that if you supply an affiliation, it has to be correct. If the signup form insists that you fill out the field, you can always write "None" or something like that. Note however opening a user account is only one part of submitting an article. In your case it is likely that you would need to be endorsed first for the category in which you want to submit. This would require you to find someone who has published a few papers in your category to endorse you. The arXiv recommends that endorsers do a quick check to see if the manuscript is appropriate for the arXiv or not. More information can be found here.

That being said, Captain Emacs makes a good point, and it would be wise to first check if your work adds anything to what is already known and available.