Cohn Measure Theory Solutions -

[ \mu(B\setminus A) = \mu(B) - \mu(A). ]

Step 2 – Necessity of finiteness. Take $X = \mathbb{R}$, $\mathcal{A} = \mathcal{B}(\mathbb{R})$ (Borel sets), $\mu = $ Lebesgue measure. Let $A = [0,\infty)$, $B = \mathbb{R}$. Then $A \subseteq B$, but $\mu(A) = \infty$. The right‑hand side $\mu(B) - \mu(A)$ is $\infty - \infty$, which is undefined in the extended real numbers. The left‑hand side $\mu(B\setminus A) = \mu((-\infty,0)) = \infty$. Thus the equality fails in the sense that the subtraction is not well‑defined. This shows $\mu(A) < \infty$ is necessary. cohn measure theory solutions

[ \mu(B) = \mu(A) + \mu(B\setminus A). ] [ \mu(B\setminus A) = \mu(B) - \mu(A)

Step 1 – Finite measure case. Since $A \subseteq B$, we have $B = A \cup (B\setminus A)$ and the union is disjoint. Finite additivity of $\mu$ (which holds for any measure) gives: Let $A = [0,\infty)$, $B = \mathbb{R}$

Subtract $\mu(A)$ from both sides (allowed because $\mu(A) < \infty$):