Given that $U_1,U_2$ are $iid$ Uniform(0,1), find the variance of $$T=\frac{\ln{U_1}}{\ln U_1 + \ln(1-U_2)}$$
Let $U\sim Uniform(0,1)=U(0,1)$
Multiply numerator and denominator by $-1$. Since $1-U\sim U(0,1)$, let $1-U_2=U_3 \sim U(0,1)$: $$T=\frac{-\ln{U_1}}{-\ln U_1 - \ln(U_3)}$$
Convert to exponential, and write the exponential as a gamma: $$T=\frac{X}{X+Y}, X, Y \sim exp(1)=Gamma(1,1)$$
Use the property that if X and Y are independent such that: $X \sim Gamma(\alpha, θ), Y \sim Gamma(β, θ)$ then $ X+Y∼Beta(α,β)$ $$T \sim Beta(1,1)$$
Since $Beta(1,1) \sim U(0,1)$ $$V(T) = \frac{(b-a)^2}{12}=\frac{(1-0)^2}{12}=\frac{1}{12}$$
Since the final distribution is standard uniform, I wanted to know if there is an easier solution that is not so circuitous