Bayesian inference is used to obtain real frequency information from imaginary-time quantum Monte Carlo data through analytic continuation. The method involves using an entropic prior and a Gaussian likelihood function. The analytic continuation problem is ill-posed, requiring regularization to handle noisy and incomplete data. The challenge lies in ensuring the data is consistent with the continuation procedure to avoid incorrect results. The method involves converting the integral equation into a linear system and using a constrained least-squares approach to maximize entropy and minimize chi-squared error. The solution requires careful statistical characterization of the data, including ensuring statistical independence and reducing variance. The results are sensitive to the quality of the data and the assumptions made. The methods allow for the determination of the spectral density from the imaginary-time data, with the accuracy depending on the consistency of the data with the assumed model. The Bayesian approach provides a framework for handling the ill-posed nature of the problem, leading to more reliable results. The methods are applicable to various quantum simulations and do not rely on specific properties of quantum Monte Carlo. The work highlights the importance of statistical consistency and the challenges in handling noisy data. The results demonstrate the effectiveness of Bayesian methods in obtaining accurate spectral densities from imaginary-time data.Bayesian inference is used to obtain real frequency information from imaginary-time quantum Monte Carlo data through analytic continuation. The method involves using an entropic prior and a Gaussian likelihood function. The analytic continuation problem is ill-posed, requiring regularization to handle noisy and incomplete data. The challenge lies in ensuring the data is consistent with the continuation procedure to avoid incorrect results. The method involves converting the integral equation into a linear system and using a constrained least-squares approach to maximize entropy and minimize chi-squared error. The solution requires careful statistical characterization of the data, including ensuring statistical independence and reducing variance. The results are sensitive to the quality of the data and the assumptions made. The methods allow for the determination of the spectral density from the imaginary-time data, with the accuracy depending on the consistency of the data with the assumed model. The Bayesian approach provides a framework for handling the ill-posed nature of the problem, leading to more reliable results. The methods are applicable to various quantum simulations and do not rely on specific properties of quantum Monte Carlo. The work highlights the importance of statistical consistency and the challenges in handling noisy data. The results demonstrate the effectiveness of Bayesian methods in obtaining accurate spectral densities from imaginary-time data.