Latent variables refer to variables that cannot be directly observed. These variable are defined in terms of observable variables.

learn more… | top users | synonyms

0
votes
0answers
32 views

The difference between SVD and SVD++

What if any is the connection between SVD (the one you learn about in your linear algebra course) and SVD++ (the one from the Netflix prize)? I know they both want to find latent factor spaces. But ...
0
votes
1answer
27 views

Compute the user and item features in SVD++

I have a sparse matrix. There is lots of missing data. Hence, I can't use SVD naively. I read Koren's SVD++ paper. I'm confused as to how the $q_i$ and $p_u$ vectors are determined. $q_i^Tp_u$ is ...
1
vote
0answers
20 views

Use of latent semantic analysis in topic modeling

I am learning Latent Semantic Analysis. After we have the SVD decomposition of term-document matrix, how to infer topics from that? For example, how to get the following result: Topic 1: apple, ...
0
votes
0answers
33 views

How to make the most of a Gaussian mixture assumption in a model?

I have a dataset with 100 columns and approximately 100000 lines. I have a variable to predict that is Y (0,1 so it's a classification problem). I have an other categorical variable with two values ...
1
vote
0answers
35 views

Log Transformation in Growth Mixture Model (Mplus user)

I am using parallel process growth mixture modeling to estimate bidirectional longitudinal dynamics between pain (0-100 theoretical range of values) and sleep measures (0-20 theoretical range of ...
4
votes
2answers
94 views

Are there any good papers comparing different philosophical views of cluster analysis?

Lots of people use cluster analysis. I've heard very few explicitly say why. I imagine this is because within a given field, most researchers seem to understand why clustering is used for the problems ...
1
vote
0answers
53 views

Variational Posterior Dirichlets in LDA

I am running the c code for LDA provided on David Blei's website. The code outputs several files. The output file final.gamma is supposed to include the "Variational Posterior Dirichlets". If I ...
0
votes
0answers
28 views

Convert latent proabilities to categorical realizations

Assume that each person can only have one of three eye colors: blue, brown or green. Furthermore assume that I have two vectors that describe for N individuals the (simulated) probability that the ...
0
votes
2answers
50 views

Can I turn a latent variable be treated as an observed variable?

I am a doctoral candidate and for my dissertation research I am using two standardized scales as independent variables. Because these have not been used with my particular population, I needed to do ...
0
votes
0answers
22 views

Categorical observed, Binomial latent

This might be a trivial problem but wondering if there's a known model formulation for this problem: Ordered Categorical observed variable with k levels Trying to estimate the parameter (theta) of a ...
1
vote
1answer
55 views

What happens when you initialize EM with a consistent estimate?

I have a certain family of models with latent random variables (not observed in the data), which I have a consistent estimator for. I now run EM on top of it, meaning, I get the consistent ...
2
votes
1answer
76 views

What are the main differences between classical and Gibbs sampling Latent Dirichlet Allocations?

In these weeks I have been studying the classical Latent Dirichlet Allocation (LDA) algorithm by David Blei and colleagues (2003), and the LDA variant based on Gibbs sampling introduced by Tom ...
0
votes
0answers
14 views

Heuristic to determine the number of latent classes [duplicate]

Do anyone know if in the field of topic modeling and latent class analysis, it exists some heuristic to determine a good number of latent classes?
1
vote
0answers
31 views

Map a normal distribution N(x,s) to an ordinal response variable

I am working on a simulation. I am going to extract a series of normally distributed values from a distribution N(x,s) whose mean x and variance s is known. I want to pretend that such distribution ...
1
vote
1answer
68 views

Latent Dirichlet Allocation - understanding the posterior

I have a problem understanding the posterior for computing LDA, stated in page 7 of Blei (2007). From my point of view, it's not exactly consistent with Bayes' theorem, as described here. Could anyone ...
2
votes
0answers
63 views

Model validation in Bayesian statistics from a model with latent variables

I am working with some two-regime autoregressive models first introduced by Hamilton in 1989. The specific models is of no great concern to my question, but some variables within my autoregressive ...
1
vote
0answers
25 views

Latent partitioning variable governing unobservable dependent variables — how to estimate?

I've got a situation where I'm willing to assume that $$ Q_t = Pe^{-k^f t} + (1-P)e^{-k^s t} $$ This is a double-exponential decay model; there is a fast pool and a slow pool. I don't know $P$, ...
0
votes
0answers
39 views

Calculating point AND uncertainty estimates for IRT factor scores

I have a survey dataset that includes items designed to measure several latent variables. There are around 3-5 items per latent variable. The items generally use 5-point response sets, so I would like ...
0
votes
0answers
63 views

Determine Relationship between Categorical & Latent Variables, running SEM Model

I have a two part question, I believe. After running some analysis, a (predictor) strategy variable in my model is (unexpectedly) categorical. I now have four categories of strategy. I need to ...
1
vote
1answer
86 views

Problems in scale Bayesian network mode using R

The problem that we have is as follows. We have close to 60 discrete random variables each of which shall take on an average of 5 categorical values. We have developed a Bayesian network ...
1
vote
0answers
66 views

Gibbs sampling in document clustering

I'm trying to understand the idea behind the Topic Models in document clustering. In Latent Dirichlet Allocation, it is necessary to approximate the posterior distribution of topics over the document. ...
1
vote
0answers
44 views

Unexpected correlation, Method to account for the shared variance

Summary: While looking at the correlations I noticed a wired, relation. Variables: User comments - A; User community activity- B; Comments on user's wall - C (control) I expected that A and B will ...
-1
votes
2answers
132 views

Any good book for learning probability programming

Are there any good books for me to learn probability programming? For example, I am new to Latent Dirichlet allocation (LDA) and Gibbs sampling. I have read some books about the techniques, but it ...
0
votes
1answer
35 views

Likelihood of a document-term matrix

I'm trying to deeply understand the Probabilistic Latent Semantic Indexing (PLSI) algorithm. The generative model underlying the algorithm is the folowing: Select a document $d \in D$ with a ...
0
votes
0answers
81 views

Toolkits for learning topic models from a co-occurrence matrix

I have a word-document co-occurrence matrix and would like to learn a topic model from it. Do you know any toolkits that take a matrix instead of a document collection as input?
1
vote
0answers
46 views

Probabilistic latent semantic analysis - objective function

In pLSA we are trying to maximize a likelihood function. The likelihood function can be given as: $L = \Pi_i^N \Pi_j^M P(d_i,w_j)^{n(d_i,w_j)}$ Where, M is the number of words and N is the number of ...
1
vote
1answer
114 views

Confusion related to EM algorithm

I was reading this tutorial related to EM algorithm at http://aass.oru.se/~tdt/ml/extra-readings/EM_algorithm.pdf. As given in the tutorial we can see that at each E step we calculate the ...
0
votes
0answers
27 views

Estimating a system of unobserved continuous stochastic processes

Say at time $t$ I have a set of observed random variables $ \{Y_i\}_{i=1}^n \subset \mathbb{R^+}$ and a set of unobserved random variables $ \{X_i\}_{i=0}^n \subset \mathbb{R^+} $ with the ...
2
votes
0answers
35 views

How to model the distribution over payouts from aggregate (sum) data?

The setting A user is presented with a number of options related to her search query. She chooses one of them. As a result, a payout is generated. The payout lies in the range [0.05, 0.50], but the ...
0
votes
0answers
165 views

Path coefficient above 1, but differential separate predictions of outcome variable.

How can two predictor latent variables have a standardized path coefficient of 1.06 (SEM: Amos: 5 indicators for the first latent variable, 3 parcel indicators for the other predictor originally from ...
2
votes
1answer
53 views

Are latent variable models modelling causality?

Is the purpose of latent variable models to model causality, where the causes are not observable i.e. latent? Are latent variables modelling causes of the observable variables? Thanks and regards!
3
votes
2answers
538 views

Expectation maximization on Bayesian networks with latent variables

I am trying to determine parameters in a bayesian network with two latent variables (in blue). Every variable is discrete with 2-4 categories. The latent variables have 3 categories each. I am ...
2
votes
1answer
95 views

Why does latent variable modelling in regression tend to push R Squared up?

Hypothetically, say I have I have three manifest variables measuring anxiety and three manifest variables (items) measuring stress. Then I want to use both to predict scores on depression, which I'm ...
0
votes
0answers
36 views

Should experimental treatments be treated as exogenous variables in structural equation modelling?

Say I have 2 experimental treatment (e.g. wearing a hat(hat/no hat) & wearing sunscreen(wearing/not wearing)) that are hypothesized to influence 2 latent variables (e.g. perceived sun protection ...
4
votes
1answer
123 views

IRT/Rasch modeling with very large N

I want to fit a 1-parameter IRT model on a questionaire with 15 questions and about six million people. Considering the large N, standard errors aren't essential. It looks like the IRT world is sort ...
0
votes
0answers
56 views

Are correlations transitive between manifested construct variables and other set of variables?

I have seen here that correlation is not transitive, but I wanted to know if such argument applies when we model our assumptions over a construct. More specifically, in my model, I have a construct ...
1
vote
0answers
34 views

Tricky latent parameter problem

I'm reinterpreting the data of some colleagues to see if I can't get at heterogeneity in grouped outcomes. Start with a decay process with two pools: $$ C_{it} = P_i e^{-k_{i}^{fast}t} + ...
1
vote
1answer
63 views

Latent variable model and graphical models

At first I was under the impression that latent variable models and graphical models were different things altogether, but after reading some papers about the former class of models, it seems they ...
1
vote
0answers
131 views

Rescale factor scores from factor analysis to latent metric in R

I'm calculating a factor analysis of several variables in R. I want to determine each case's value on the latent variable. When I run the factor analysis, I receive factor scores. The factor scores do ...
1
vote
1answer
114 views

fractions or wholes for the slope in a latent growth model?

When trying to model a latent growth model in lavaan and AMOS respectively, they seem to approach the time spacing of the slope estimates differently. Lavaan defaults to whole numbers increasing by ...
3
votes
1answer
262 views

Will a quadratic and cubic latent growth curve model always have better fit than a linear model?

A conjecture that I have heard is this: When you add a quadratic or a cubic term to a linear latent growth curve model, the fit will always improve. Is this correct? Why/why not? As I have had ...
1
vote
1answer
671 views

High SRMR despite good fit based on other indices in SEM (latent growth curve)

Context: Latent Growth Curve Modeling for continuous variable with 15 time points. One intercept, two slopes (for first and ...
4
votes
1answer
57 views

Latent variables in Bayes nets with no physical interpretation

In Pattern Recognition and Machine Learning Bishop writes about Bayes networks: For practical applications of probabilistic models, it will typically be the highernumbered variables ...
2
votes
1answer
100 views
5
votes
2answers
676 views

Which R package to use to conduct a latent class growth analysis (LCGA) / growth mixture model (GMM)?

I am trying to perform a latent class growth analysis (LCGA) and/or growth mixture models (GMMs) in R. The data I am using is an increasing number of forks of git repositories (discrete variable, not ...
3
votes
0answers
52 views

Can we reconstruct the hidden (latent) variables after executing EM?

The question is in the title. I know that EM algorithm could do maximum likelihood estimation for models that have latent variables. I would like to know can we get the (estimated) value of these ...
1
vote
1answer
169 views

Confusion related to calculation of conditional distribution

I have this confusion related to the calculation of a conditional distribution suppose $y_n = N(0,w)$ $p(o_n|y_n) = N(D.y_n,\phi)$ How do I calculate $p(y_n|o_n)$ Actually I was reading this ...
1
vote
0answers
408 views

Entering panel-data (cross-sectional time-series data) into SPSS

I don't know how to enter pooled data (I have data about 42 countries over 7 years per variable) into SPSS and then run a regression. So far, SPSS does not assign the observations to a certain ...
1
vote
2answers
92 views

Latent Variables

Suppose $Y$ is an ordinal variable such that $Y = 1,2,3,4$ corresponds to levels of impairment. So $Y=1$ is the last impaired and $Y = 4$ is the most impaired. What is the purpose of latent variables? ...
3
votes
1answer
177 views

Rule of thumb reliability of scales for using manifest variables in path analysis?

I am using path analysis to test hypotheses in my study. I have been told by a few people in passing that I can use manifest variables rather than latent variables as long as my reliabilities are ...