German Stata Users Group Meeting: Announcement and Program
Date, Place and Costs
June 13, 2014
University of Hamburg
Van-Melle-Park 6 ("Philosophenturm")
Hörsaal E (ground flour)
http://www.uni-hamburg.de/index_e.html
Meeting only: 45 Euro (students 25 Euro)
Workshop on June 12: 65 Euro
Workshop and Conference: 85 Euro
(including coffee, tee, luncheons)
Overview
Meeting: The 12th German Stata Users Group Meeting will be held at the the University of Hamburg on Friday, June 13 2014. Everybody from anywhere who is interested in using Stata is invited to attend this meeting. The meeting will include presentations about causal models, general statistics, and data management, both by researchers and by StataCorp staff. The meeting will also include a "wishes and grumbles" session, during which you may air your thoughts to Stata developers.
Workshop: On the day before the conference, Ulrich Kohler, co-author of the book "Data Analysis Using Stata" and several user written Stata commands will present a workshop on "Visualization of Statistical Data Using Stata". Details about the workshop are given below the program.
Conference Dinner: There is (at additional cost) the option of an informal meal at a restaurant in Hamburg on Friday evening. Details will be given at the conference.
Language: The conference language will be English because of the international nature of the meeting and the participation of non-German guest speakers.
Time table
Abstracts
8:30-9:00 Registration
9:00-9:15 Welcome
9:15-10:15 Regression analysis of censored data using pseudo-observations
Erik T. Parner, Aarhus University,Denmark
[email protected]
Abstract: In a series of papers, a method based on pseudo-values has been proposed for direct regression modeling of the survival function, the restricted mean and cumulative incidence function in competing risks with right censored data. The models, once the pseudo-values have been computed, can be fitted using standard generalized estimating equation software. In this talk I will present three Stata procedures to compute these pseudo-observations, and give examples of applications. Guidelines for the number of variables in the regression analyses are presented, and future updates of the procedures are discussed.
10:15-11:15 Using Stata for Sequence Analysis
Brandan Halpin, University of Limerick
[email protected]
Abstract: Sequence analysis is a very different way of looking at categorical longitudinal data, such as life-course or labour-market histories (or any ordered categorical data, for that matter). Instead of focusing on transition rates (e.g., via hazard rate, Markov or panel models) it takes individual time-series and compares them as wholes. It has significant advantages at a descriptive and exploratory level, and may help detect patterns which conventional methods will overlook. As availability of longitudinal data increases, this becomes a significant advantage.
Sequence analysis hinges on defining measures of similarity between sequences, typically in order to generate data-driven classifications, for example by cluster analysis. Most SA uses the Optimal Matching distance, but others measures are in use. There is some controversy about the applicability of SA algorithms to social science data, and about their parameterisation. Comparison of different methods and parameterisations helps clarify the issues.
For a long time TDA was the only package social scientists had access to for sequence analyis, but in more recent years both Stata and R have had relevant functionality, in Stata's case provided by the SQ and SADI packages. [BR] In this talk I will discuss the current state of the SADI package. SADI differs from SQ in being plugin-based, and therefore is significantly faster: many of the distance measures are compute-intensive, and typically N*(N-1)/2 comparisons will be made. It also provides additional distance measures, including Dynamic Hamming, Time-Warp Edit Distance and a version of Elzinga's Number of Matching Subsequences measure. It includes tools for inspecting and graphing sequence data, and for comparing distance measures and the resulting cluster analyses.
I will also briefly discuss the advantages and disadvantages of using plugins rather than Mata, and make some remarks about cross-compiling plugins under Linux.
11:15-11:30 Coffee
11:30-12:30 Some examples using gsem to handle endogeneity in nonlinear models
David M. Drukker, StataCorp
[email protected]
Abstract: Unobserved components can parameterize problems of endogeneity in many nonlinear models for cross-sectional and panel data. This talk provides some examples and uses -gsem- to estimate the parameters.
12:30-13:30 Lunch
13:30-14:00 Managing Stata related files with dirtools
Ulrich Kohler, University of Potsdam
[email protected]
Abstract: The presentation will illustrate severals uses of the programs in the dirtools package. dirtools is a collection of programs designed to deal with native Stata files (i.e. .dta, .do, .ado, .mata, or .gph) and some of the more frequently generated other file formats (.eps, .pdf, .tex). The programs provides an easy access to typical tasks to be done with these file types. The user can, for example, describe or load datasets, compile mata files, translate .gph files to .eps or .pdf files, compile TeX files, etc. In addition the package allows an easy way to change the working directory and to define bookmarks for frequently used directories.
14:00-14:30 Splitting spells in very large or daily datasets
Klaudia Erhardt, DIW Berlin (SOEP) and Ralf Künstler, WZB (NEPS)
[email protected],[email protected]
Abstract: We present two Stata Programs dealing with spell data: ``splitspells.do'' and ``combispells.do''. ``splitspells.do'' is a syntax to split spell data so that every splitted spell is either completely parallel or completely unique to other spells within the same case. This could be done by splitting each spell in single time-unit splits, a method which is not recommendable - or even feasible - for large datasets or daily data, as it produces a maximum of additional records. In contrast, ``splitspells'' does the job by producing the least possible number of additional records, and is a useful tool in the process of transforming multiple spelltype-datasets into ``unidimensional'' sequence data.
The second program, ``combispells'' is to be used with already splitted episode data. Using existing spelltype-variables and user-defined short labels it produces a (labelled) numeric variable as well as a string variable that show the spelltyp-combinations occurring within parallel splits of a case. Thus it provides a user friendly and easy-to-handle tool for the edition, revision, and exploration of spell data. Both amply commented programs are available at the authors.
As ``combispells'' is easy to understand on one's own account, the main focus of the presentation will be on ``splitspells''.
14:30-15:00 Modeling Interactions in Count Data Regression. Principles and Implementation in Stata
Heinz Leitgöb, University of Linz, Austria
[email protected]
Abstract: During the past decades, count data models (in particular Poisson and Negative Binomial based regression models) have gained relevance in empirical social research. While identifying and interpreting main effects is relatively straightforward for this class of models, the integration of interactions between predictors proves to be complex. As a consequence of the exponential mean function implemented in count data models (restricting the possible range of the conditional expected count to nonnegative values), the coefficient of the product term variable (generated by the predictors constituting the interaction) does - in contrast to the linear model - not fully represent the underlying interaction effect. Further, the interaction effect is allowed to vary between individuals and can be divided into two components: (i) a model inherent interaction effect and (ii) a product-term induced interaction effect.
We will derive the total interaction effect for the Poisson and Negative Binomial models by following a method developed by Norton and Ai (Economics Letters, 80 (2003) 123-129) for binary Logit and Probit models. Further, we will decompose the model inherent and the product-term induced interaction effect, discuss their substantive meaning and provide Delta method standard errors for the respective effects. Finally, an approach for the estimation and graphical representation of these effects in Stata will be provided in detail.
15:00-15:15 Coffee
15:15-16:15 Estimating average treatment effects from observational data using teffects
David M. Drukker, StataCorp
[email protected]
Abstract: After reviewing the potential-outcome framework for estimating treatment effects from observational data, this talk discusses how to estimate the average treatment effect and the average treatment effect on the treated using the regression-adjustment estimator, the inverse-probability-weighted estimator, two doubly robust estimators, and two matching estimators implemented in teffects.
16:15-16:45 Reproducible Research in Stata
Bill Rising, StataCorp
[email protected]
Abstract: Writing a document that contains statistical results in its narrative, including inline results, can take too much effort. Typically, writers have a separate series of do-files whose results must then be pulled into the document. This is a very high-maintenance fashion to work in because updates to the data, changes to the do-files, updates to the statistical software, and, especially, updates to inline results all require work and careful checking of results.
Reproducible research greatly lessens document-maintenance chores by putting code and results directly into the document; this means that only one document need be maintained; thus it remains consistent and is easily maintained.
In this presentation, I will show you how to put Stata code directly into a LaTeX or HTML document and run it through a preprocessor to create the document containing results. While this is useful for creating self-contained documents, it is tremendously useful for creating periodic reports, class notes, solution sets, and other documents that get used over a long period of time.
16:45-17:00 Coffee
17:00-17:30 A new command for plotting regression coefficients and other estimates
Ben Jann, University of Bern, Switzerland
[email protected]
Abstract: Graphical display of regression results has become increasingly popular in presentations and the scientific literature, as graphs are much easier to read than tables in many cases. In Stata such plots can be produced by the -marginsplot command. However, while marginsplot is very versatile and flexible, it has two major limitations: it can only process results left behind by margins and it can only handle one set of results at the time. In this presentation I will introduce a new command called coefplot that overcomes these limitations. It plots results from any estimation command and combines results from several models into a single graph. The default behavior of coefplot is to plot markers for coefficients and horizontal spikes for confidence intervals. However, coefplot can also produce various other types of graphs. The capabilities of coefplot are illustrated using a series of examples.
17:30-18:15 Report to the Users and wishes and grumbles
Bill Rising, StataCorp and Users
[email protected]
Abstract: A Stata developer talks about developments at Stata and users
talk about how Stata should develop.
Workshop: Visiulisation of Statistical Data Using Stata
by Ulrich Kohler
[email protected]
Thursday, June 12 2014, 9:15-17:00
University of Hamburg
Van-Melle-Park 9
WP A514 (PC-Pool)
[URL]http://www.uni-hamburg.de/index_e.html[URL]
Presenter Prof. Kohler holds the chair for Methods of Empirical Social Research
at the University of Potsdam. He is co-author of the book "Data
Analysis Using Stata" and author of several user written Stata
commands.
Costs 65 Euro (including coffee, tea, luncheons)
(Workshop and Conference: 85 Euro)
Register For registration please email to Anke Mrosek ([email protected])
Contents The visualization of statistical data has become an increasingly important tool for researcher to analyze their data and present their findings. This workshop aims at providing participants with the basics of graphical perception theories and familiarize them with the manifold options to create graphs with Stata. The workshop covers arbitrary graphs using the so called "resultsset-approach" and introduces the power of marginsplot for showing the results of complicated regression models.
Registration and accommodations
Please travel at your own expense. The conference fee will be 45 (Students 25). There will also be an optional informal meal at a restaurant in Hamurg on Friday evening at additional cost. You can enroll by contacting Anke Mrosek ([email protected]) by email or by writing, phoning, or faxing to
Anke Mrosek
Dittrich Partner Consulting GmbH
Prinzenstrasse 2
42697 Solingen
Germany
Tel: +49 (0)212 260 6624
Fax: +49 (0)212 260 6666
Scientific Organizers
The academic program of the meeting is being organized by D. Enzmann, K.-U. Schnapp (University of Hamburg), J. Giesecke (Humboldt University Berlin) and U. Kohler (University of Potsdam).
Logistics organizers
The logistics are being organized by Dittrich and Partner (http://www.dpc-software.de), the distributor of Stata in several countries including Germany, The Netherlands, Austria, Czech Republic, and Hungary.
Date, Place and Costs
June 13, 2014
University of Hamburg
Van-Melle-Park 6 ("Philosophenturm")
Hörsaal E (ground flour)
http://www.uni-hamburg.de/index_e.html
Meeting only: 45 Euro (students 25 Euro)
Workshop on June 12: 65 Euro
Workshop and Conference: 85 Euro
(including coffee, tee, luncheons)
Overview
Meeting: The 12th German Stata Users Group Meeting will be held at the the University of Hamburg on Friday, June 13 2014. Everybody from anywhere who is interested in using Stata is invited to attend this meeting. The meeting will include presentations about causal models, general statistics, and data management, both by researchers and by StataCorp staff. The meeting will also include a "wishes and grumbles" session, during which you may air your thoughts to Stata developers.
Workshop: On the day before the conference, Ulrich Kohler, co-author of the book "Data Analysis Using Stata" and several user written Stata commands will present a workshop on "Visualization of Statistical Data Using Stata". Details about the workshop are given below the program.
Conference Dinner: There is (at additional cost) the option of an informal meal at a restaurant in Hamburg on Friday evening. Details will be given at the conference.
Language: The conference language will be English because of the international nature of the meeting and the participation of non-German guest speakers.
Time table
| 8:30-9:00 | Registration | |
| 9:00-9:15 | Welcome | |
| 9:15-10:15 | Regression analysis of censored data using pseudo-observation |
Erik T. Parner |
| 10:15-11:15 | Using Stata for Sequence Analysis | Brendan Halpin |
| 11:15-11:30 | Coffee | |
| 11:30-12:30 | Some examples using gsem to handle endogeneity in nonlinear models |
David Drukker |
| 12:30-13:30 | Lunch | |
| 13:30-14:00 | Managing Stata related files with dirtools |
Ulrich Kohler |
| 14:00-14:30 | Spellsplitting using very large or daily datasets | Klaudia Erhardt, Ralf Künstler |
| 14:30-15:00 | Modeling Interactions in Count Data Regression | Heinz Leitgöb |
| 15:00-15:15 | Coffee | |
| 15:15-16:15 | Esitmating average treatment effect from observational data using teffects |
David Drukker |
| 16:15-16:45 | Reproducible research in Stata | Bill Rising |
| 16:45-17:00 | Coffee | |
| 17:00-17:30 | A new command for plotting regression coefficients and other estimates |
Ben Jann |
| 17:30-18:15 | Report to the Users/Whishes and Grumbles | Bill Rising, Users |
8:30-9:00 Registration
9:00-9:15 Welcome
9:15-10:15 Regression analysis of censored data using pseudo-observations
Erik T. Parner, Aarhus University,Denmark
[email protected]
Abstract: In a series of papers, a method based on pseudo-values has been proposed for direct regression modeling of the survival function, the restricted mean and cumulative incidence function in competing risks with right censored data. The models, once the pseudo-values have been computed, can be fitted using standard generalized estimating equation software. In this talk I will present three Stata procedures to compute these pseudo-observations, and give examples of applications. Guidelines for the number of variables in the regression analyses are presented, and future updates of the procedures are discussed.
10:15-11:15 Using Stata for Sequence Analysis
Brandan Halpin, University of Limerick
[email protected]
Abstract: Sequence analysis is a very different way of looking at categorical longitudinal data, such as life-course or labour-market histories (or any ordered categorical data, for that matter). Instead of focusing on transition rates (e.g., via hazard rate, Markov or panel models) it takes individual time-series and compares them as wholes. It has significant advantages at a descriptive and exploratory level, and may help detect patterns which conventional methods will overlook. As availability of longitudinal data increases, this becomes a significant advantage.
Sequence analysis hinges on defining measures of similarity between sequences, typically in order to generate data-driven classifications, for example by cluster analysis. Most SA uses the Optimal Matching distance, but others measures are in use. There is some controversy about the applicability of SA algorithms to social science data, and about their parameterisation. Comparison of different methods and parameterisations helps clarify the issues.
For a long time TDA was the only package social scientists had access to for sequence analyis, but in more recent years both Stata and R have had relevant functionality, in Stata's case provided by the SQ and SADI packages. [BR] In this talk I will discuss the current state of the SADI package. SADI differs from SQ in being plugin-based, and therefore is significantly faster: many of the distance measures are compute-intensive, and typically N*(N-1)/2 comparisons will be made. It also provides additional distance measures, including Dynamic Hamming, Time-Warp Edit Distance and a version of Elzinga's Number of Matching Subsequences measure. It includes tools for inspecting and graphing sequence data, and for comparing distance measures and the resulting cluster analyses.
I will also briefly discuss the advantages and disadvantages of using plugins rather than Mata, and make some remarks about cross-compiling plugins under Linux.
11:15-11:30 Coffee
11:30-12:30 Some examples using gsem to handle endogeneity in nonlinear models
David M. Drukker, StataCorp
[email protected]
Abstract: Unobserved components can parameterize problems of endogeneity in many nonlinear models for cross-sectional and panel data. This talk provides some examples and uses -gsem- to estimate the parameters.
12:30-13:30 Lunch
13:30-14:00 Managing Stata related files with dirtools
Ulrich Kohler, University of Potsdam
[email protected]
Abstract: The presentation will illustrate severals uses of the programs in the dirtools package. dirtools is a collection of programs designed to deal with native Stata files (i.e. .dta, .do, .ado, .mata, or .gph) and some of the more frequently generated other file formats (.eps, .pdf, .tex). The programs provides an easy access to typical tasks to be done with these file types. The user can, for example, describe or load datasets, compile mata files, translate .gph files to .eps or .pdf files, compile TeX files, etc. In addition the package allows an easy way to change the working directory and to define bookmarks for frequently used directories.
14:00-14:30 Splitting spells in very large or daily datasets
Klaudia Erhardt, DIW Berlin (SOEP) and Ralf Künstler, WZB (NEPS)
[email protected],[email protected]
Abstract: We present two Stata Programs dealing with spell data: ``splitspells.do'' and ``combispells.do''. ``splitspells.do'' is a syntax to split spell data so that every splitted spell is either completely parallel or completely unique to other spells within the same case. This could be done by splitting each spell in single time-unit splits, a method which is not recommendable - or even feasible - for large datasets or daily data, as it produces a maximum of additional records. In contrast, ``splitspells'' does the job by producing the least possible number of additional records, and is a useful tool in the process of transforming multiple spelltype-datasets into ``unidimensional'' sequence data.
The second program, ``combispells'' is to be used with already splitted episode data. Using existing spelltype-variables and user-defined short labels it produces a (labelled) numeric variable as well as a string variable that show the spelltyp-combinations occurring within parallel splits of a case. Thus it provides a user friendly and easy-to-handle tool for the edition, revision, and exploration of spell data. Both amply commented programs are available at the authors.
As ``combispells'' is easy to understand on one's own account, the main focus of the presentation will be on ``splitspells''.
14:30-15:00 Modeling Interactions in Count Data Regression. Principles and Implementation in Stata
Heinz Leitgöb, University of Linz, Austria
[email protected]
Abstract: During the past decades, count data models (in particular Poisson and Negative Binomial based regression models) have gained relevance in empirical social research. While identifying and interpreting main effects is relatively straightforward for this class of models, the integration of interactions between predictors proves to be complex. As a consequence of the exponential mean function implemented in count data models (restricting the possible range of the conditional expected count to nonnegative values), the coefficient of the product term variable (generated by the predictors constituting the interaction) does - in contrast to the linear model - not fully represent the underlying interaction effect. Further, the interaction effect is allowed to vary between individuals and can be divided into two components: (i) a model inherent interaction effect and (ii) a product-term induced interaction effect.
We will derive the total interaction effect for the Poisson and Negative Binomial models by following a method developed by Norton and Ai (Economics Letters, 80 (2003) 123-129) for binary Logit and Probit models. Further, we will decompose the model inherent and the product-term induced interaction effect, discuss their substantive meaning and provide Delta method standard errors for the respective effects. Finally, an approach for the estimation and graphical representation of these effects in Stata will be provided in detail.
15:00-15:15 Coffee
15:15-16:15 Estimating average treatment effects from observational data using teffects
David M. Drukker, StataCorp
[email protected]
Abstract: After reviewing the potential-outcome framework for estimating treatment effects from observational data, this talk discusses how to estimate the average treatment effect and the average treatment effect on the treated using the regression-adjustment estimator, the inverse-probability-weighted estimator, two doubly robust estimators, and two matching estimators implemented in teffects.
16:15-16:45 Reproducible Research in Stata
Bill Rising, StataCorp
[email protected]
Abstract: Writing a document that contains statistical results in its narrative, including inline results, can take too much effort. Typically, writers have a separate series of do-files whose results must then be pulled into the document. This is a very high-maintenance fashion to work in because updates to the data, changes to the do-files, updates to the statistical software, and, especially, updates to inline results all require work and careful checking of results.
Reproducible research greatly lessens document-maintenance chores by putting code and results directly into the document; this means that only one document need be maintained; thus it remains consistent and is easily maintained.
In this presentation, I will show you how to put Stata code directly into a LaTeX or HTML document and run it through a preprocessor to create the document containing results. While this is useful for creating self-contained documents, it is tremendously useful for creating periodic reports, class notes, solution sets, and other documents that get used over a long period of time.
16:45-17:00 Coffee
17:00-17:30 A new command for plotting regression coefficients and other estimates
Ben Jann, University of Bern, Switzerland
[email protected]
Abstract: Graphical display of regression results has become increasingly popular in presentations and the scientific literature, as graphs are much easier to read than tables in many cases. In Stata such plots can be produced by the -marginsplot command. However, while marginsplot is very versatile and flexible, it has two major limitations: it can only process results left behind by margins and it can only handle one set of results at the time. In this presentation I will introduce a new command called coefplot that overcomes these limitations. It plots results from any estimation command and combines results from several models into a single graph. The default behavior of coefplot is to plot markers for coefficients and horizontal spikes for confidence intervals. However, coefplot can also produce various other types of graphs. The capabilities of coefplot are illustrated using a series of examples.
17:30-18:15 Report to the Users and wishes and grumbles
Bill Rising, StataCorp and Users
[email protected]
Abstract: A Stata developer talks about developments at Stata and users
talk about how Stata should develop.
Workshop: Visiulisation of Statistical Data Using Stata
by Ulrich Kohler
[email protected]
Thursday, June 12 2014, 9:15-17:00
University of Hamburg
Van-Melle-Park 9
WP A514 (PC-Pool)
[URL]http://www.uni-hamburg.de/index_e.html[URL]
Presenter Prof. Kohler holds the chair for Methods of Empirical Social Research
at the University of Potsdam. He is co-author of the book "Data
Analysis Using Stata" and author of several user written Stata
commands.
Costs 65 Euro (including coffee, tea, luncheons)
(Workshop and Conference: 85 Euro)
Register For registration please email to Anke Mrosek ([email protected])
Contents The visualization of statistical data has become an increasingly important tool for researcher to analyze their data and present their findings. This workshop aims at providing participants with the basics of graphical perception theories and familiarize them with the manifold options to create graphs with Stata. The workshop covers arbitrary graphs using the so called "resultsset-approach" and introduces the power of marginsplot for showing the results of complicated regression models.
- Graphical Perception
- An introductory experiment
- Pattern perception
- Table lookup - Plotting regression models
- Understanding margins
- marginsplots
- Common problems, proper solutions - Create arbitrary graphs with the resultsset-approach
- Limitations of charts
- graph twoway
- Tipps and tricks with graph twoway
Registration and accommodations
Please travel at your own expense. The conference fee will be 45 (Students 25). There will also be an optional informal meal at a restaurant in Hamurg on Friday evening at additional cost. You can enroll by contacting Anke Mrosek ([email protected]) by email or by writing, phoning, or faxing to
Anke Mrosek
Dittrich Partner Consulting GmbH
Prinzenstrasse 2
42697 Solingen
Germany
Tel: +49 (0)212 260 6624
Fax: +49 (0)212 260 6666
Scientific Organizers
The academic program of the meeting is being organized by D. Enzmann, K.-U. Schnapp (University of Hamburg), J. Giesecke (Humboldt University Berlin) and U. Kohler (University of Potsdam).
Logistics organizers
The logistics are being organized by Dittrich and Partner (http://www.dpc-software.de), the distributor of Stata in several countries including Germany, The Netherlands, Austria, Czech Republic, and Hungary.
