For testing of the fairness of a coin using Bayesian approach, why non-informativs prior is a good choice?

18 Views Asked by At

The probablity density for tossing fair coin question is $f(x|θ)=θ(1−θ)^{x−1}$, $x\in\{1,2,...\}$ where $θ$ is the probability of throwing a head. We allocate $θ$ a flat prior, $π(θ) = 1$ and test is carried out $14$ times.

Why non-informative prior(beta(1,1)) is reasonable?

Is this because the limit of beta prior is improper?(does not integrate to 1)