Questions tagged [expectation-maximization]

Question 1

I have data from a survey. This data includes a key metric of interest (Y), which has an interesting distribution: a clear peak at 0, and a generally right-skewed distribution. I have modeled the ...

Question 2

A common example that I have found for explaining expectation-maximization is the example of two biased coins. The problem statement is: You have two biased coins, which you select with equal ...

Question 3

I am quite new to ML methods such as GMMs but I have a problem at hand which requires me to estimate the covariance matrices of Gaussians such that the datapoints are drawn from a weighted sum of ...

Question 4

Learning about EM algorithms and finite mixture models and I've run into a particularly unintuitive problem. I'm trying to fit a finite mixture regression model on simulated data, where the true ...

Question 5

Inlined at the bottom is the code of a MATLAB simulation I wrote. This code very simply runs expectation maximization for three Gaussians and, as set down, is supposed to illustrate the degeneracy ...

Question 6

I have a dataset of (noisy) test results. You can think of this as being accuracy that a chess player achieves in various games or number of points a basketballer scores in a game. I think a good ...

Question 7

I'm reading a paper on applying a SLOPE model to a Bayesian spike-and-slab framework [link]. The SLOPE penalty is given by $$ \text{pen}(\lambda) = \sigma \sum_{j=1}^{p} \lambda_{r(\beta,j)} |\beta_j|,...

Question 8

Let $\sigma>0$. Suppose we observe $N$ samples of sizes $T_1,\dots,T_n$ that are each generated by the following data generating process: $\theta_n$ is drawn from the distribution $\mathrm{N}(0,\...

Question 9

I am trying to use DLM package in R to estimate a state space model, where the observation and state equation are as follows: $y_t=\beta_1a_t + B_t\beta_2(\frac{u_t-v_t}{u_t}) + C_t\beta_3(\frac{u_t-...

Question 10

I am reading the paper Some contributions to maximum likelihood factor analysis. Consider the factor analysis model $$y=\Lambda x+z, $$ where $y$ is a vector containing $p$ features and $x$ is the ...

Question 11

I'm reading about how the estimation of the survival function is "self-consistent" because as Efron showed, we can estimate the survival function in the presence of right censoring as $$ \...

Question 12

These questions arose when I was reading Online Appendix D for the paper Missing Events in Event Studies: Identifying the Effects of Partially-Measured News Surprises by R.S. Gurkaynak, B. Kisacikoglu ...

Question 13

I have a general idea of Gaussian Mixture Models. My understanding: GMM is a way of clustering data points which, unlike K means clustering, soft assigns them under different distributions by ...

Question 14

From cs 229 page 6: Intuitively, the EM algorithm alternatively updates Q and θ by a) setting Q(z) = p(z|x; θ) following Equation (8) so that ELBO(x; Q, θ) = log p(x; θ) for x and the current θ, and ...

Question 15

ELBO is a lower bound, and only matches the true likelihood when the q-distribution/encoder we choose equals to the true posterior distribution. Are there any guarantees that maximizing ELBO indeed ...

Stack Exchange Network

Questions tagged [expectation-maximization]

Extending mixture distribution (EM algorithm) for analysis

Substituting per-trial missing data for all missing data in the two-coin expectation-maximation example

Parameter estimation using a GMM with negative weights

Why do overfitted models in finite mixture regression sometimes have the smallest BIC despite the true number of components being selected frequently?

Concrete example of degenerate expectation maximization - why does it differ from my intuition?

Inferring the parameters of a Ornstein-Uhlenbeck process given realisations of it cannot be observed directly

Understanding this penalised complete-data log-likelihood

Consistency of classical estimation of hyper-parameters from multiple samples

Estimating state space model with exogenous states

How do we guarantee the result is the global minimizer when dealing a function related to MLE of factor analysis?

On self-consistent estimators of survival functions. How do they work?

On a non-standard application of Kalman filter

How to perform user assisted image segmentation using Gaussian Mixture Models?

In Expectation-Maximization, in the maximization step, do we maximize expectation of the log likelihood (wikipedia) or evidence lower bound (cs 229)?

Why does Variational Inference work?

Hot Network Questions