A new iterative method for calculating complex posterior distributions using data augmentation
In short: Researchers Tanner and Wong developed an iterative algorithm to estimate posterior distributions that are otherwise difficult to calculate directly. By augmenting observed data with latent, or missing, values, the researchers demonstrated that one can approximate complex distributions by averaging results from multiple imputations of the missing data.
The method addresses situations where likelihoods cannot be approximated by a normal distribution, making standard error estimates unreliable. The algorithm works by alternating between sampling from the augmented data posterior and the predictive distribution of the latent data. While the authors provide a way to approximately sample parameters for certain multinomial submodels, the complexity of sampling from these distributions can still present implementation challenges.
Why it matters: This mathematical framework provides a way to perform Bayesian inference even when the underlying data is incomplete or the distributions are non-normal.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: Research: Tanner and Wong (1987). The Calculation of Posterior Distributions by Data Augmentation. Journal of the American Statistical Association.
To cite this card, link to https://intellectplus.co.uk/app/science-research-a-new-iterative-method-for-calculating-complex-posterior-dis.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Science · Latest · A–Z · All topics