- #1
friend
- 1,452
- 9
Suppose we have an observable with a certain number of eigenstates. We would normalize all these possibilities to 1 in order to give each eigenstate an appropriate probability of being measured. Can we then only consider the data of many measurements for only a subset of those eigenstates and normalize that subset to 1 and get different probabilities for considering only that subset of alternatives? Is that subset called a subspace of the original Hilbert space? And can this be done for any arbitrary subset of the original eigenstates?
Last edited: