All posts by Margot Sijssens-Bennink

FAQ Latent GOLD®

General


What
resources are available to learn about Latent GOLD® and latent class
modeling?


What
data file formats can Latent GOLD® handle?


How can I use Latent GOLD&reg with SAS data sets?


How
many records and variables can I use? How much time will it take to
run?


How
does Latent GOLD® differ from the LEM Program?

Do you have any tutorials for event history analysis?

LC Cluster Analysis


How
does latent class cluster analysis compare with the traditional clustering
procedures in SAS and SPSS?


How
does Latent GOLD® classify cases into latent classes?


When the ‘Include Missing’ option is selected, does Latent Gold do some kind of imputation?

How
are Latent Class (LC) clustering techniques related to Fuzzy Clustering
Techniques?

How can I tell if my latent class cluster model contains local dependencies?

How can I handle local dependencies in my LC cluster model?

In LC cluster models containing continuous indicators, how can I
determine whether a model should contain a within class correlation
between 2 or more of these variables?

LC Factor Analysis


How
does latent class factor analysis compare with traditional factor
analysis?


Latent class
factor analysis running time.


LC Regression Analysis


How
does LC regression analysis compare with traditional regression
modeling?


I
need a mixture modeling program that can handle dependent variables that are
dichotomous as well as continuous. Does Latent GOLD® handle this?


I have a binary dependent variable and five categorical independent variables. I am using Latent GOLD® to find 3 segments among the respondents. The Parameters output shows separate estimates for each segment. However, there appears to be both intercepts as well as betas for dependent variable. I am confused about how to use both of them in terms of predicting.


Is
there any “stepwise” inclusion feature in the LC regression
module?


Can
Latent GOLD® perform multinominal LC regression models?  Can it be used with
repeated measures such as obtained in conjoint and discrete choice
studies?


For LC Regression models, there are several R square statistics reported in the Latent Gold output. When there are 2 or more latent segments (latent classes), do these still measure the overall strength of the predictors to predict the dependent variable?


I understand that the covariates are used to predict membership in a class based upon the probabilities derived from a multinomial logit model. The classification errors, reduction errors, entropy R square, etc. are associated with this estimation. Correct?

Questions
on Tutorial #3: LC Regression with Repeated Measures


After I run the model, say on a binary
response, and get two latent classes with their set of parameters, I’d
like to predict the response of a new observation, with a given set of
predictors and with or without covariates, but unknown response. How can
I get this from Latent Gold?


Technical Questions from Latent GOLD®
Users


I
scored my data file using the ‘classification output file’ option and found that
the percentage of each class is different than the class sizes given in the
profile output
.


I
have included several ordinal variables with many values in my model and the
program takes a very long time to run.  Can I do anything to speed it
up?


The
output listings in the manual for the IRIS data contain some errors in the
statistics.  What are the correct results?”


Latent GOLD Advanced Questions


What additional functionalities are gained with the advanced version of Latent Gold?


I have the Advanced version of Latent GOLD. Is it possible to
estimate IRT models such as the Rasch model, Rost’s Rasch mixture model,
partial credit model and rating scale models that can be estimated in the
WINMIRA program? If so, how do the parameterizations differ?


LG-Syntax Questions


How does Latent GOLD takes into account the complex sampling scheme during multiple imputation? Is there some technical documentation that explains the details?


Replication and Case Weights


I’ve been reading through pp. 56-57 of the Latent Gold manual to try to understand the difference between ‘Replication Weights’ and ‘Case Weights,’ and I’m having some difficulty understanding which I should be using. I have survey data where different respondents may have differing number of rows of data. Different respondents will generally have different weights, but the weight variable would have the same value for each row of data within a respondent. The respondent weights reflect how much we want that respondent to ‘count’ in any analysis. It’s not clear to me whether I should import the weight variable into ‘Replication Weight’ or ‘Case Weight’ when I’m setting up my analysis. I want to use all the data – i.e., it’s not a question of creating a holdout sample by weighting down certain rows to 1.0e-100.


 


General


Q.
What resources are available to learn about Latent GOLD® and latent class
modeling?


A. Before purchasing the program, you can try out the free
demo
version
of the program, which allows access to all program features with
sample data files. Tutorials
take you step-by-step through several analyses of these sample files. These
tutorials along with several articles
are available on our website. Upon purchase of the program users can download a 200
page User’s Guide that covers a wide range of topics on Latent Class Analysis
and Latent GOLD® . We also offer a once a year training
program
(Statistical Modeling Week) which includes a 2 day course on Latent
Class Analysis, as well as Online Courses


Q. What data file
formats can Latent GOLD® handle?


A. Latent GOLD® can handle ASCII
Text data formats as well as SPSS files.

Return to
List of Questions


Q. How can I
use Latent GOLD® with SAS data sets?


SAS Export can create an SPSS .sav file which can be opened by Latent GOLD®. The SAS Documentation
illustrates the Export function. View the Relevant Export page for instructions.

Q.
How many records and variables can I use? How much time will it take to
run?


A .There is NO limit concerning the number of records. The time will depend on several factors including
the # of variables and records, speed of your machine, and the requested output.
For many models, Latent GOLD® runs 20 or more times faster than other Latent
Class programs. We suggest trying the demo program to see how fast Latent GOLD®
works on your machine.


Q. How does Latent
GOLD differ from the LEM program?


Latent GOLD® implements the 3 most important types of
latent class (LC) models. It was designed to be extremely easy to use and to
make it possible for people without a strong statistical background to apply LC
analysis to their own data in a safe and easy way. LEM is a command language
research tool that Prof. Jeroen Vermunt developed for applied researchers
with a strong statistical background who want to apply nonstandard
log-linear and latent class models to their categorical data. With LEM you can
specify more probability structures with many more kinds of restrictions (if you
know how to do it), but is not designed to be Windows friendly, requires strict
data and input formats and does not provide error checks.

With Latent
GOLD, continuous and count variables can be included in the model, and special
LC output not available in LEM is provided, such as various graphs,
classification statistics, and bivariate residuals. Latent GOLD® also has faster
(full Newton-Raphson) and safer (sets of starting values, Bayes constants)
estimation methods for LC models than LEM. Both programs give information on
nonidentifiability and boundary solutions, but Latent GOLD® , unlike LEM, can
prevent boundary solutions through the use of Bayes constants.

Q. Do you have any tutorials for event history analysis?


The set of example data files on our website contains various
event history analysis examples. Tutorials are not yet available for
these. However, to get you started, you might look at the data file land.sav, the full reference for which is ” Land, K.C., Nagin, D.S., and
McCall (2001). Discrete-time hazard regression
models with hidden heterogeneity: the
semi-parametric mixed Poisson approach.
Sociological Methods and Research, 29,
342-373.” Another good example is
jobchange.dat.



Land.sav contains information on 411 males
from working-class area of London who were
followed from ages 10 through 31. The
dependent variable is “first serious
delinquency”.
As can be seen, there is one record for
each time point, which is called a person-
period data format. The dependent “first” is
zero for all records of a person, expect for
the last if a person experienced the event of
interest at that age.



The variables age and age_sq are the
duration variables. These can also be seen
as time-varying predictors. The variable “tot”
is a time-constant covariate/predictor (a
composite risk factor).
Of course the ID should be used as Case ID to
indicate which records belong to the same
case.



The dependent “first” can be treated as a
Poisson count or as a binomial count. The
former option yields a piece-wise constant log-
linear hazard model, the latter a discrete-time
logit. If treated as Poisson count, it is best to
set the exposure to one half (exp_half: event
occurs in the middle of the interval) for the
time point at which the event occurs. With a
binomial count the exposure should be one all
the time (=default).
Age and age_sq should be used as class-
dependent predictors. You identify two
groups with clearly different age pattern in the
rate of first delinquency. The variable “tot” can
be used as class-independent predictor, but
more interesting is to use it as covariate: does
the risk factor determine the type of
delinquency trajectory?



This example can be modified.extended in
many ways.
– you can include other time-varying predictors
than the time variables. These can be
assumed to have the same or different effects
across classes.
– you can include information on another
event. In that case your classes describe the
pattern in multiple events
– you can include as many covariates as you
want (this will usually be demographics, but
can also be a treatment)
– you can model the time dependence as
nominal, yielding a Cox-like model.



A general reference on event history
combined with LC analysis is Vermunt (1997),
Log-linear event history analysis. Sage
Publications..

Return to
List of Questions

LC Cluster Analysis


Q.
How does latent class cluster analysis compare with the traditional clustering
procedures in SAS and SPSS?


A. LC clustering is model-based in contrast to traditional
approaches that are based on ad-hoc distance measures. The general probability
model underlying LC clustering more readily accommodates reality by allowing for
unequal variances in each cluster, use of variables with mixed scale types, and
formal statistical procedures for determining the number of clusters, among many
other improvements. For a detailed comparison showing how LC cluster outperforms
SPSS K-means clustering and SAS FASTCLUS procedures, see Latent Class
Modeling as a Probabilistic Extension of K-means Clustering
.

Published article on a comparison of SPSS TwoStep Cluster with Latent GOLD


Q. How does
Latent GOLD® classify cases into latent classes
.


A. Cases are assigned to the latent class having the
highest posterior membership probability. Covariates can be added to the model
for improved description and prediction of the latent classes.


Return to
List of Questions


Q.
When the ‘Include Missing’ option is selected, does Latent Gold do some kind of imputation?


A. No, imputation is not necessary. Classification with missing values works exactly the same as classification without
missing values. It is simply based on the variables that are observed for the case
concerned. There is no imputation of missing values for indicators. One of the nice
things about LC analysis is that imputation is not necessary.


In the User’s Guide, we give the general form of the density with missing
values. The crucial thing is the delta, which is 0 if an indicator is missing. If that occurs the
term cancels (it is equal to 1 irrespective of the value of y).

Thus with 4 indicators y1, y2, y3, and y4, two clusters and y2 missing

P(x|y1,y3,y4) = P(x) P(y1|x) P(y3|x) P(y4|x) / P(y1,y3,y4)

where

P(y1,y3,y4) = P(1) P(y1|1) P(y3|1) P(y4|1) + P(2) P(y1|2) P(y3|2) P(y4|2)


Return to
List of Questions

Q.
How are Latent Class (LC) clustering techniques related to Fuzzy Clustering
Techniques


A. In fuzzy clustering, a case has grades of
membership
which are the “parameters” to be estimated (Kaufman and
Rousseeuw, 1990). In contrast, in LC clustering an individual’s posterior
class-membership probabilities are computed from the estimated model parameters
and the observed scores. The advantage of the LC approach is that it is possible
to use the LC model to classify other cases (outside the sample used to
estimate the model) which belong to the population. This is not possible with
standard fuzzy clustering techniques.


Kaufman, L. and Rousseeuw, P.J. 1990. Finding groups in
data: An introduction to cluster analysis, New York: John Wiley and
Sons.


Return to
List of Questions

Q.
How can I tell if my latent class cluster model contains local dependencies?


A. Local dependence for a K-class model exists if the model does NOT fit the
data. One such measure of model fit is given by the bivariate residuals (BVRs)
associated with each pair of model indicators. Large BVRs (values over 2)
can be viewed as evidence of local dependence
associated with that pair of model indicators (see the
residuals output of Latent GOLD for these statistics).


Return to
List of Questions

Q.
How can I handle local dependencies in my LC cluster model?


A. Local dependence can
be accounted for by simply adding latent classes or by maintaining the
current number of classes and modifying the model in other ways such as
adding direct effects associated with 2 variables that have large bivariate
residuals. See the LG manual for details of how to add
direct effects. See also section 3 in
http://www.statisticalinnovations.com/articles/sage11.pdf for further
details of the different approaches for dealing with local dependence.


Return to
List of Questions

Q.
In LC cluster models containing continuous indicators, how can I
determine whether a model should contain a within class correlation
between 2 or more of these variables?


A. You can estimate several models and select the one that fits best according to BIC.
For example, six types of LC cluster models are reported in Table 1 of the Latent Class
Cluster Analysis article
. These models differ with respect to a) the specification of class dependent vs. class independent error variances and
b) the ‘direct effects’ included in the LC cluster model estimated by
LatentGOLD. The 3-class type-5 model is best according to the BIC statistic.
Various parameter estimates and standard errors from this ‘final’ model
are obtained from the Profile and Parameters Output.

Click on dataset #29
and download the data and the model setup file diabetes.lgf containing the specifications for each of the 6 types of 3-class cluster models described in Table 1.


Return to
List of Questions

LC Factor Analysis


Q.
How does latent class factor analysis compare with traditional factor analysis?


A. The LC factor model assumes that each factor contains 2
or more ordered categories as opposed to traditional factor analysis which
assumes that the factors (as well as the variables) are continuous (interval
scaled). The variables in LC factor analysis need not be continuous. They may be
mixed scale types (nominal, ordinal, continuous, counts, or combinations of
these). LC Factor also has a close relationship to cluster analysis. For an
introduction to LC factor analysis, and to see how it relates to LC cluster
analysis, see Magidson and
Vermunt Sociological Methodology 2001
. For a comparison with traditional
Factor Analysis in datamining see Traditional vs. Latent
Class Factor Analysis for Datamining


Q. I was
looking at the lifestyle data set (tutorial #1) and was trying to run a factor
model on all lifestyle indicator variables (from Tennis to Military). I have
requested an 8-factor model and it has been running for 30 minutes. Am I doing
something wrong?


A. A 2-factor model on a 650 MH computer took less than 2
minutes to estimate and a 3-factor model 4 minutes. As the number of factors
increases the estimation time increases exponentially. From an exploratory
perspective, you may well find that a 2 or 3 factor solution will already be
quite informative — 3 dichotomous factors will segment the sample into 8
distinct clusters! On the other hand, 8 dichotomous factors corresponds to 2 to
the power 8 = 256 clusters. To see the relationship between factors and
clusters, see Magidson and
Vermunt Sociological Methodology 2001
.


Traditional factor analysis (FA) is faster because it makes a simplifying
assumption that all variables are continuous and that they follow a multivariate
normal (MVN) distribution. When these assumptions are true, only the second
order moments (the correlations between the variables) are needed to estimate
the model. For these data, the FA assumptions are not justified.


Latent GOLD® does not assume MVN and hence is much more general. It utilizes
information from all higher order associations (more than means and
correlations) in the estimation of parameters. The resulting solution will be
directly interpretable and unique, unlike the traditional FA solution which
requires a rotation for interpretability. Traditional vs. Latent
Class Factor Analysis for Datamining
is an article by the developers of LG
that will appear in a book on datamining. It shows why LG factor analysis often
provides insights into data that are missed by traditional FA.


Return to
List of Questions


LC Regression Analysis


Q.
How does latent class regression analysis compare with traditional regression
modeling?


A. There are 2 primary kinds of differences. First, the
particular regression is automatically determined according to the scale
type of the dependent variable. For continuous, the traditional linear
regression is employed; for dichotomous, logistic regression; for ordinal, the
baseline/adjacent category logit extension; for nominal, multinomial logit; for
count, Poisson regression. models are used. For example, for dichotomous
dependent variables, the logistic regression model is used. Second, LC
Regression is a mixture model and hence is more general than traditional
regression. The special case of 1-class corresponds to the homogeneous
population assumption made in traditional regression. In LC regression, separate
regressions are estimated simultaneously for each latent class.


I
need a mixture modeling program that can handle dependent variables that are
dichotomous as well as continuous. Does Latent GOLD® handle this?


A. Yes. Mixture modeling and latent class
modeling are synonymous


Return to
List of Questions


Is
there any “stepwise” inclusion feature in the LC regression
module?


A. No. Since the latent classes may be highly dependent on
the predictors that are included, stepwise features have not been implemented in
the latent class regression module.

Q.
I have a binary dependent variable and five categorical independent variables. I am using Latent GOLD® to find 3 segments among the respondents. The Parameters output shows separate estimates for each segment. However, there appears to be both intercepts as well as betas for dependent variable. I am confused about how to use both of them in terms of predicting.


A. The ‘gamma’ parameters labeled ‘Intercept’ (and other gamma parameters that would appear if you have covariates) refer to the model to predict the latent variable classes as a function of the covariates. If no covariates are included in the model only the Intercept appears under the label ‘(gamma)’. Beneath the gamma parameters, the parameters labeled ‘beta’ appear. These refer to the model to predict the dependent variable (which including the dependent variable regression intercept). This output has been rearranged in Latent GOLD® to provide better separation of the parameters from these two different models. See Tutorial 3 (PDF) for an example.

Latent GOLD® also has many additional features useful for prediction, such as
the automatic generation of predicted values, the ability to restrict the regression coefficients in many ways, and R-square statistics. See the User’s Guide (PDF) and Technical Guide(PDF) for further details on these new features.

Can
Latent GOLD® perform multinominal LC regression models?  Can it be used with
repeated measures such as obtained in conjoint and discrete choice
studies?


A. Multinomial LC regression models are estimated simply
by specifying the dependent variable to be nominal. In the case of repeated
measures, (multiple time points, multiple ratings by the same respondent, etc.)
an ID variable can be used to identify the records associated with the same
case. (See tutorial
#2
for an example of a repeated measures conjoint study.) Latent GOLD®
cannot currently estimate conditional logit models of the kind used in
discrete choice studies, although such capability will be incorporated in Latent
GOLD Choice, and add-on to Latent GOLD® , that is now under
development.

For LC Regression models, there are several R square statistics reported in the Latent Gold output. When there are 2 or more latent segments (latent classes), do these still measure the overall strength of the predictors to predict the dependent variable?


A. Yes. One important additional aspect is that estimated class-membership
also improves overall prediction and contributes to the magnitude of R square.

I understand that the covariates are used to predict membership in a class based upon the probabilities derived from a multinomial logit model. The classification errors, reduction errors, entropy R square, etc. are associated with this estimation. Correct?


A. This is not fully correct: These measures indeed indicate how well we can predict class membership. But, the covariates alone do not determine classification — the regression model itself plays a major role in predicting class membership. This prediction/classification is based on a person’s responses on the dependent variable (given predictor values). If you look at the formulas, you can see that the posterior membership probabilities do not only depend on P(x|z), but also on P(y|x,z). Even without any covariates (z), these models usually predict class membership quite well.

Intuitively, one determines which class-specific regression model fits best to the responses of a certain case. The better that a regression model associated with a particular class fits, the higher the probability of belonging to that class. Price sensitive people are assigned to the class for which the regression shows higher price effects, etc.

In Latent GOLD, we also report a separate R-squared for the prediction of class membership based on covariates only.


Return to
List of Questions


Q.
I need additional information on Tutorial # 3: LC Regression with Repeated
Measures. Specifically, I would like to know how preference ratings and
probabilities for different preference levels for a given profile are computed
in this example. In other words, how do I use the estimated beta coefficients to
compute the probabilities of choosing Ratings 1 thru 5 for a given profile, say
[Fashion=Traditional, Quality=High, Price=Lower]? Also, what exactly are the
gamma coefficients as distinguished from those labeled betas in
the parameters output?


A. The ‘ordinal’ dependent variable specification is used
in this example which causes the baseline category logit model to be used. The
beta coefficients listed in the column of the parameters output
file
corresponding to a particular latent class are the b-coefficients in
the following model:


f( j | Z1, Z2, Z3) = b0(j) + b1*Z1*y(j) + b2*Z2*y(j) +
b2*Z3*y(j).


The b0 estimates are the betas associated with each rating
category j of the dependent variable RATING.


The y(j), j=1,2,3,4,5 are the fixed scores used for the
dependent variable, (1, 2, 3, 4, and 5 in this example)


The desired probabilities are thus computed as:


Prob(Rating = j | Z1, Z2, Z3) = exp[f(j)] /
[exp(f(1))+exp(f(2))+exp(f(3))+exp(f(4))+exp(f(5))] , j = 1,…,5


(For additional technical information on this model see
the associated Magidson references)


“Maximum Likelihood Assessment of Clinical Trials Based on
an Ordered Categorical Response.”  Drug Information Journal, Maple
Glen, PA: Drug Information Association, Vol. 30, No. 1, 1996.


“Multivariate Statistical Models for Categorical Data,”
chapter 3 in Bagozzi, Richard, Advanced Methods of Marketing Research,
Blackwell, 1994.


Additional coefficients, labeled gammas (as
opposed to betas) pertaining to the multinomial logit model for
predicting the latent variable as a function of the covariates (SEX and
AGE for this example) are listed at the bottom of the parameters output file (in Latent GOLD®). In
the model containing no covariates, the gamma coefficients (labeled
‘intercepts’) relate to the size of the classes which are always ordered from
largest (the first latent class) to smallest (the last class).


Return to
List of Questions


Q.
After I run the model, say on a binary
response, and get two latent classes with their set of parameters, I’d
like to predict the response of a new observation, with a given set of
predictors and with or without covariates, but unknown response. How can
I get this from Latent Gold?


A. With active covariates, posterior membership probabilities are computed
for
cases with missing responses (whether or not they are ‘new’ cases), based on
their covariate values, as shown in Latent GOLD’s ‘covariate classification’
output. These probabilities are used as weights applied to the predictions
for each latent class, using the predictors for such cases and the
regression coefficients associated with that class to get the appropriate
prediction for each class. Without active covariates, the posterior
membership probabilities are taken to be the overall class sizes.

In practice, if the new cases are included in the data file but given a case
weight close to 0 (say 1E-49), while all other cases are given a weight of
1, and the
‘include missing’ option is used, such cases will not be used to estimate
the model parameters (so the same solution would be obtained without the new
cases), but by requesting that Predictions be output to a file, predictions
for ALL cases, including the new cases, will be output to the file..


Return to
List of Questions


Technical Questions from Latent GOLD®
Users


Q.
I scored my data file using the ‘classification output file’ option and found
that the percentage of each class is different than the class sizes given in the
profile output. What am I (or Latent GOLD® ) doing wrong?


A. Nothing is wrong. What you are observing is the effects
of misclassification errors associated with assignment to a latent class based
upon the modal (highest) class probability. For example, in a 3-class model if
the posterior membership probabilities for cases having a given response pattern
are 0.2 (for class 1), 0.7 (for class 2), and 0.1 (for class 3), the modal
probability is 0.7. Assignment based on the modal probability means that
all such cases will be assigned to class 2. However, such assignment is
expected to be correct for only 70% of these cases, since 20% truly belong to
class 1 and the remaining 10% belong to class 3. The expected misclassification
rate for these cases will be 20% + 10% = 30%. For cases with other response
patterns, the misclassification rate may be 7%, or 2%, etc. The modal assignment
rule minimizes the overall expected misclassification rate (the
overall expected misclassification rate is given in the output). To the
extent the misclassification rate is greater than 0, the observed frequency
distribution of class memberships will reflect the effects of such
misclassification. The marginal distributions in the classification table show how the marginal distribution changes when using modal class assignment.


Return to
List of Questions


Q.
I have included several ordinal variables with many values in my model and the
program takes a very long time to run? Can I do anything to speed it
up?


A. Substantial improvement in speed can be accomplished by
specifying the variables to be continuous. Alternatively, the grouping option
can be used to reduce the number of levels in the ordinal variables (to say 10
or 20).


Q.
The output listings in the manual for the IRIS data contain some errors in the
statistics. What are the correct results?


A. The output listings in the manual for the IRIS data
contain some errors in the statistics. You can download the correct
specification for each model (iris.lgf)
and the data
(iris.dat)
.

Return to
List of Questions


Latent GOLD Advanced Questions


Q. What additional functionalities are gained with the advanced version of Latent Gold?


A. The advanced version of Latent GOLD consists of an advanced module containing the ability to 1) estimate multi-level latent class models, 2) incorporate complex sampling designs, and 3) include random effects with continuous factors (CFactors). An overview of these capabilities is provided in section 8 of the Technical Guide, followed by detailed documentation for each of these 3 advanced features in sections 9, 10 and 11, as well as output produced by these advanced features in section 12. You may download a demo version that contains the Advanced module and use it with any of our demo data sets.


Q. I have the Advanced version of Latent GOLD. Is it possible to
estimate IRT models such as the Rasch model, Rost’s Rasch mixture model,
partial credit model and rating scale models that can be estimated in the
WINMIRA program? If so, how do the parameterizations differ?


A. Yes, Latent GOLD Advanced (LGA) can be used to estimate a wide variety of
IRT and IRT mixture models. This .pdf describes the connections
between various LGA and standard IRT models. The associated .lgf and data
files
illustrate examples that can be run with our demo data sets. (Note
that we set the Bayes constant to 0 in these runs.)


Return to
List of Questions


LG-Syntax Questions


Q. How does Latent GOLD takes into account the complex sampling scheme during multiple imputation? Is there some technical documentation that explains the details?


Yes it does. This is not documented specific to multiple imputation, but the nonparametric
bootstrap used in the multiple imputation procedure (to account for parameter uncertainty) deals with
the complex sampling design (the resampling is done at the PSU level). This is explained
in the LG-Syntax User’s Guide when discussing complex sampling features. On page 85 we say: “In a
nonparametric bootstrap, replicate samples are obtained by sampling C(o) PSUs with
replacement from each stratum. The values of delta(r) is 1/R.”


Return to
List of Questions


Replication and Case Weights


Q. I’ve been reading through pp. 56-57 of the Latent Gold manual to try to understand the difference between ‘Replication Weights’ and ‘Case Weights,’ and I’m having some difficulty understanding which I should be using. I have survey data where different respondents may have differing number of rows of data. Different respondents will generally have different weights, but the weight variable would have the same value for each row of data within a respondent. The respondent weights reflect how much we want that respondent to ‘count’ in any analysis. It’s not clear to me whether I should import the weight variable into ‘Replication Weight’ or ‘Case Weight’ when I’m setting up my analysis. I want to use all the data – i.e., it’s not a question of creating a holdout sample by weighting down certain rows to 1.0e-100.


A. In this situation you should use a case weight since you want to modify the weights of cases.

A replication weight is used to increase or decrease the weight of a
choice within a case. A weight of ‘2’ means that this person made this choice twice.
So setting all replication weights to 2 for a particular case is not the
same as having a case weight of 2:

  • A case weight of 2 means: there are 2 cases with this set of choices (I have to count
    this person twice).

  • A series of replication weights of 2 means: this person made each of these choices twice (I have
    twice as much information for one person).

In the special case of a 1 class model, the two are equivalent because all observations are assumed to be independent. In all other situations, they are very different. You can consult the manual to see how the weights enter into the log-likelihood function, which is very
different for the two types of weights.

Presentation: A New Modeling Tool for Identifying Meaningful Segments and their Willingness to Pay: Improving Validity by Reducing the Confound between Scale and Preference Heterogeneity

Conference:
2015 Advanced Research Techniques (ART) Forum

Date:
June 15, 2015

Presenter(s):
Jay Magidson, Statistical Innovations

Location:
San Diego, CA

Abstract:
With discrete choice data, it is common practice to use HB or latent class modeling to capture heterogeneity across decision-makers. However, a significant part of the heterogeneity retrieved relates to differences in the amount of error variance, a phenomenon referred to as scale heterogeneity. As a result of the preference vs. scale heterogeneity confound, traditional approaches to segmentation may yield spurious segments that differ only in their scale heterogeneity, and may not differ at all in their willingness to pay. In this presentation we introduce a new scale adjusted latent class (SALC) choice model that accounts for both observed and unobserved scale heterogeneity, resulting in homogeneous segments that differ only in their preferences and willingness to pay, increasing the usefulness of segmentation analyses to marketers. We compare results from the SALC approach with other standard and nonstandard approaches.

More information and registration:
https://www.ama.org/events-training/Conferences/Pages/Advanced-Research-Techniques-%28ART%29-Forum.aspx

 

Course: Latent Class Models for Multilevel and Longitudinal Data

Date:
July 8-10, 2015

Teacher(s):
Jeroen K. Vermunt, Tilburg University
Margot Sijssens-Bennink, Statistical Innovations

Location:
Barcelona , Spain

Course content:
This course deals with various more advanced application types of latent class (LC) analysis. These concern applications with multilevel and longitudinal data sets. More specifically, you will learn how to use LC regression models, LC growth models, latent Markov models, and multilevel LC models. During this could we will use the Latent GOLD computer program, including the Syntax module.

First we will look into the data organization for these more advanced LC analysis applications. These are univariate or multivariate two-level data files, which contain multiple records per unit, with a single or multiple response variables and in most cases also predictor variables.

The LC or mixture regression model is a two-level regression model in which the intercept and predictor effects are allowed to vary across units by assuming that units belong to latent classes. This model can also be seen as a semi-parametric variant of the standard random effects regression model. Application types include standard multilevel regression analysis and the modeling of data from rating or choice-based conjoint experiments.

One special type of application of the LC regression model involves its use with longitudinal data sets. Note that these are also two-level data sets, with time variables as the main predictors. Such models are referred to as LC growth or LC trajectory models. LC growth models may contain continuous random effects in addition to latent classes.

The latent Markov or latent transition model is an extension of the simple LC cluster model for use with longitudinal data. Latent Markov models differ from LC growth models in that they can also be used with multiple responses and in that respondents are allowed to change their class membership over time. Basically, we model the transition across latent states. Extension of the basic model include models with covariates and models with a mover-stayer structure.

The multilevel LC model is an extension of the simple LC cluster model to the situation in which lower-level units are nested within higher-level units. In this model not only individuals are clustered into a small number of classes, but also the groups are clustered into group-level classes.

Requirement for successful participation in this course are basic knowledge of LC analysis (for example, the introductory course of the summer school), multilevel regression analysis, and longitudinal data analysis.

More information and registration:
http://www.upf.edu/survey/Summer/

Course: Introduction to Latent Class Cluster Analysis

Date:
July 6-8, 2015

Teacher(s):
Jeroen K. Vermunt, Tilburg University
Margot Sijssens-Bennink, Statistical Innovations

Location:
Barcelona , Spain

Course content:
Latent class (LC) analysis is used in a broad range of research fields with the aim to cluster respondents into a small number subgroups called latent classes. Typically the observed variables used in a LC analysis are categorical variables, but it also possible to use continuous variables or counts. In this introductory course you will learn how to perform such an analysis by yourself.  We will not only focus on practical skills, but also discuss the most important parts of the underlying theory. All steps will be illustrated using empirical examples. During this course we will use the software package Latent GOLD.

First, using a simple example application, we present the structure of the basic LC model for categorical variables and discuss its key model assumptions. Subsequently, we switch to a more realistic example to discuss the important issue of model selection or, more specifically, how to determine the number of latent classes. For model selection, we use information criteria (AIC, AIC3, and BIC), goodness-of-fit statistics (possibly with bootstrap p-values), bivariate residuals, bootstrapped likelihood-ratio tests, and Wald tests.

Once a model with a specific number of latent classes is selected, respondents are classified into the latent classes based on their posterior class membership probabilities. We discuss how these classifications are obtained, as well as how the quality of the classifications can be evaluated using so-called classification statistics.

Subsequently, we will discuss various important extensions of the basic LC model, such as LC models which relax the local dependence assumption, and LC models for ordinal, continuous, and count variables.

Finally, we deal with the important issue on how to investigate the relationship between the individuals’ class memberships with covariates and distal outcomes. This can be done in various ways, among others using the recently developed bias adjusted three-step approach. We also discuss multiple-group LC analysis.

You don’t need any background in LC analysis to be able to follow this introductory course. Also researchers with some experience in LC analysis will benefit from reviewing the basics since they will pick up more of the details while beginning users can focus on the global learning goals of the course.

More information and registration:
http://www.upf.edu/survey/Summer/