Archive

Posts Tagged ‘data science’

Stata 17 released

We just announced Stata 17. Visit stata.com/new-in-stata to read all about its 29 major new features.

They are

Looking over this list of features, someone suggested that a potential marketing tagline for Stata 17 could be “Better. Faster. Stronger.” I thought Daft Punk might not like it if we used that, so we aren’t, but really, it is a great overall description of the new version.

I’ll share my thoughts on some of the new features below.

 

Customizable tables

 
There has been a long tradition in the Stata user community of commands that build various tables. These are among some of the most-used community-contributed commands! There has been an equally long tradition of the Stata user community asking us to provide more official features to assist with flexible table creation and export. Stata’s table command has been completely revamped, and a new collect command allows you to gather and manage results from multiple commands, which can then be shown in tabular form. Excel, HTML, LaTeX, Markdown, PDF, Stata SMCL, Word, and plain text are supported as export formats. I suspect almost all users will be adding this to their Stata repertoire.

 

Bayesian econometrics

 
In Stata 17, we have added many features for Bayesian econometrics, including

Many of you have been asking us for Bayesian VAR models. That’s not surprising. VAR models have many parameters but often not enough data to estimate them reliably. The Bayesian approach provides a solution by incorporating specialized priors to allow you to obtain more stable parameter estimates.

As with classical VAR models, you can perform IRF analysis and obtain dynamic forecasts but now within the Bayesian paradigm.

Bayesian panel-data models are appealing when you have few panels or when you would like to study and compare panel-specific effects.

With Bayesian DSGE models, prior distributions give you a natural way to incorporate knowledge about model parameters that is motivated by the economic theory.

As with Stata’s other Bayesian features, our aim is to make specification of these models as intuitive as possible and as similar to the specifications of the frequentist counterparts as possible.

All the above is in addition to other existing Bayesian features of interest to econometricians such as Bayesian generalized linear models and Bayesian sample-selection models.

 

Bayesian multilevel models

 
Stata users span many disciplines. In addition to the new Bayesian features above that will be of most interest to econometricians, Stata 17 also adds Bayesian multilevel modeling with support for nonlinear, joint, SEM-like, and even more models. One notable feature is the ability to fit multivariate nonlinear models containing random effects such as multivariate nonlinear growth models.

 

Give me more speed

 
As computer capabilities have grown, so have dataset sizes. With larger datasets and more computationally intensive methods comes the above request, “Give me more speed.” Sometimes, our developers say, “I’m giving her all she’s got, Captain!”, but for Stata 17, they gave us all more.

We achieved speed gains in part through careful algorithm selection and implementation and in part through integration of the Intel Math Kernel Library (MKL) to underpin many of Mata’s linear algebra functions and operators.

Read the details, including tables of specific speed gains.

 

Difference-in-differences (DID) models

 
DID and difference-in-difference-in-differences (DDD) models are appealing to many disciplines, including econometrics, epidemiology, political science, public policy, and many more. If you are studying the effect of a treatment (such as a drug regimen or policy) in observational data and are concerned that the effect may be influenced by time or by some other group effects, DID and DDD models provide intuitive methods to control for such unobserved effects.

 

New meta-analysis features

 
Stata 17 adds the following to the excellent (IMHO) meta-analysis suite introduced in Stata 16:

If you recognize the above, you know you want them!

And we have made these new features just as easy to use as the rest of the meta suite.

 

Interval-censored Cox model

 
The Cox proportional hazards model is used routinely by researchers from many disciplines to analyze right-censored event-time data, where time to an event of interest is observed exactly. In Stata 17, you can use it with interval-censored event-time data too!

With interval-censored event-time data, we only know that the time to an event of interest lies in an interval. For instance, think of time to cancer recurrence or to an infection or, for that matter, to any asymptomatic disease that can be detected only through periodic examinations. These are all interval-censored data, and you now have a new powerful tool to analyze them.

 

New lasso features

 
Stata 17 adds the following to the popular lasso features first introduced in Stata 16:

In fact, treatment-effects lasso combines two popular features: treatment effects and lasso. You can now incorporate many (hundreds, thousands, and more) covariates in your treatment-effects models.

You can account for clustered observations in your lasso analysis. And you can use the BIC criterion to select lasso penalty parameters.

 

Integration with other software and languages

 
Two of Stata’s great features are its extensibility and reproducibility. Stata 17 builds on that tradition by greatly enhancing its interoperability with Python and Java, adding support for Jupyter Notebook, adding JDBC support, and giving you experimental access to the H2O platform.

You could already call Python code from Stata code. Now you can call Stata from any stand-alone Python environment. Do you write Python code in Jupyter Notebook, Spyder IDE, or PyCharm IDE? Now you can call Stata directly from those environments, passing data, metadata, and results between Stata and Python seamlessly. Even your Stata graphs will show up directly in Jupyter Notebook. Our nickname for all the ways you can connect Python and Stata is ‘PyStata‘.

For Java, you could already compile Java code into .jar files and call those as plugins from Stata. Now you can embed Java code directly in your Stata do-files and ado-files, just like Mata code and Python code. Stata will compile and execute your Java code on the fly. You can interchange data, metadata, and results at will.

One of the first steps of any analysis is importing your data. Stata 17 supports JDBC for importing data from and writing data to databases that provide JDBC drivers. An important advantage JDBC has over ODBC is that JDBC drivers are platform independent, so if a database vendor provides a JDBC driver, it will work seamlessly on Windows, Mac, and Linux.

Finally, some of our developers have been experimenting with connecting Stata to H2O, a scalable and distributed open-source machine-learning and predictive analytics platform. We decided to release our experiment to you. Is this something you’d like us to do more with? We look forward to your feedback.

 

New date and time functions

 
Stata 17 adds a plethora of date and time convenience functions in three main areas:

  • Datetime durations, such as ages
  • Relative dates, such as the next birthday relative to a reference date
  • Datetime components, functions that extract various components from datetime values

You’ll undoubtedly find these make your life easier when working with date and time values.

 

New Do-file Editor features

 
Don’t miss the enhancements in the Do-file Editor, including persistent bookmarks that are saved with your code, a Navigation Control providing quick access to those bookmarks and defined programs, syntax highlighting support for Java and XML (in addition to the existing support for Stata ado, Python, and Markdown), and autocompletion of quotes, parentheses, and brackets around a selection.

 

And there’s more

 
Health scientists who deal with ordinal outcomes with an overabundance of values in the lowest category will want to try the new ziologit command.

Those of you interested in panel data and categorical outcomes will be pleased to know that you can now analyze both together easily with the new xtmlogit command.

And for those of you interested in nonparametric tests of trend, we have added three new tests to the existing nptrend command.

Finally, Stata 17 runs fully natively on Apple’s new M1 Macs, known as Apple Silicon. Stata ships as a universal application that has everything necessary to run natively on both M1 Macs and Intel-based Macs for the best performance no matter your choice of hardware platform.

It has been a lot of fun to see this release come together at StataCorp, and it is a tremendous pleasure to be able to release it to you.

Stata/Python integration part 6: Working with APIs and JSON data

Data are everywhere. Many government agencies, financial institutions, universities, and social media platforms provide access to their data through an application programming interface (API). APIs often return the requested data in a JavaScript Object Notation (JSON) file. In this post, I will show you how to use Python to request data with API calls and how to work with the resulting JSON data. Read more…

Categories: Programming Tags: , , , ,

Stata 16 Released

We just announced the release of Stata 16. It is now available. Click to visit stata.com/new-in-stata.

Stata 16 is a big release, which our releases usually are. This one is broader than usual. It ranges from lasso to Python and from multiple datasets in memory to multiple chains in Bayesian analysis.

The highlights are listed below. If you click on a highlight, we will spirit you away to our website, where we will describe the feature in a dry but information-dense way. Or you can scroll down and read my comments, which I hope are more entertaining even if they are less informative.

The big features of Stata 16 are

  1. Lasso, both for prediction and for inference
  2. Reproducible and automatically updating reports
  3. New meta-analysis suite
  4. Revamped and expanded choice modeling (margins works everywhere)
  5. Integration of Python with Stata
  6. Bayesian predictions, multiple chains, and more
  7. Extended regression models (ERMs) for panel data
  8. Importing of SAS and SPSS datasets
  9. Flexible nonparametric series regression
  10. Multiple datasets in memory, meaning frames
  11. Sample-size analysis for confidence intervals
  12. Nonlinear DSGE models
  13. Multiple-group IRT
  14. Panel-data Heckman-selection models
  15. NLMEs with lags: multiple-dose pharmacokinetic models and more
  16. Heteroskedastic ordered probit
  17. Graph sizes in inches, centimeters, and printer points
  18. Numerical integration in Mata
  19. Linear programming in Mata
  20. Do-file Editor: Autocompletion, syntax highlighting, and more
  21. Stata for Mac: Dark Mode and tabbed windows
  22. Set matsize obviated

Number 22 is not a link because it’s not a highlight. I added it because I suspect it will affect the most Stata users. It may not be enough to make you buy the release, but it will half tempt you. Buy the update, and you will never again have to type

. set matsize 600

And if you do type it, you will be ignored. Stata just works, and it uses less memory.

Oh, and in Stata/MP, Stata matrices can now be up to 65,534 x 65,534, meaning you can fit models with over 65,000 right-hand-side variables. Meanwhile, Mata matrices remain limited only by memory.

Here are my comments on the highlights.

1. Lasso, both for prediction and for inference

There are two parts to our implementation of lasso: prediction and inference. I suspect inference will be of more interest to our users, but we needed prediction to implement inference. By the way, when I say lasso, I mean lasso, elastic net, and square-root lasso, but if you want a features list, click the title.

Let’s start with lasso for prediction. If you type

. lasso linear y x1 x2 x3 ... x999

lasso will select the covariates from the x‘s specified and fit the model on them. lasso will be unlikely to choose the covariates that belong in the true model, but it will choose covariates that are collinear with them, and that works a treat for prediction. If English is not your first language, by “works a treat”, I mean great. Anyway, the lasso command is for prediction, and standard errors for the covariates it selects are not reported because they would be misleading.

Concerning inference, we provide four lasso-based methods: double selection, cross-fit partialing out, and two more. If you type

. dsregress y x1, controls(x2-x999)

then, conceptually but not actually, y will be fit on x1 and the variables lasso selects from x2-x999. That’s not how the calculation is made because the variables lasso selects are not identical to the true variables that belong in the model. I said earlier that they are correlated with the true variables, and they are. Another way to think about selection is that lasso estimates the variables to be selected and, as with all estimation, that is subject to error. Anyway, the inference calculations are robust to those errors. Reported will be the coefficient and its standard error for x1. I specified one variable of special interest in the example, but you can specify however many you wish.

2. Reproducible and automatically updating reports

The inelegant title above is trying to say (1) reports that reproduce themselves just as they were originally and (2) reports that, when run again, update themselves by running the analysis on the latest data. Stata has always been strong on both, and we have added more features. I don’t want to downplay the additions, but neither do I want to discuss them. Click the title to learn about them.

I think what’s important is another aspect of what we did. The real problem was that we never told you how to use the reporting features. Now we do in an all-new manual. We tell you and we show you, with examples and workflows. Here’s a link to the manual so you can judge for yourself.

3. New meta-analysis suite

Stata is known for its community-contributed meta-analysis. Now there is an official StataCorp suite as well. It’s complete and easy to use. And yes, it has funnel plots and forest plots, and bubble plots and L’Abbé plots.

4. Revamped and expanded choice modeling (margins works everywhere)

Choice modeling is jargon for conditional logit, mixed logit, multinomial probit, and other procedures that model the probability of individuals making a particular choice from the alternatives available to each of them.

We added a new command to fit mixed logit models, and we rewrote all the rest. The new commands are easier to use and have new features. Old commands continue to work under version control.

margins can now be used after fitting any choice model. margins answers questions about counterfactuals and can even answer them for any one of the alternatives. You can finally obtain answers to questions like, “How would a $10,000 increase in income affect the probability people take public transportation to work?”

The new commands are easier to use because you must first cmset your data. That may not sound like a simplification, but it simplifies the syntax of the remaining commands because it gets details out of the way. And it has another advantage. It tells Stata what your data should look like so Stata can run consistency checks and flag potential problems.

Finally, we created a new [CM] Choice Modeling Manual. Everything you need to know about choice modeling can now be found in one place.

5. Integration of Python with Stata

If you don’t know what Python is, put down your quill pen, dig out your acoustic modem and plug it in, push your telephone handset firmly into the coupler, and visit Wikipedia. Python has become an exceedingly popular programming language with extensive libraries for writing numerical, machine learning, and web scraping routines.

Stata’s new relationship with Python is the same as its relationship with Mata. You can use it interactively from the Stata prompt, in do-files, and in ado-files. You can even put Python subroutines at the bottom of ado-files, just as you do Mata subroutines. Or put both. Stata’s flexible.

Python can access Stata results and post results back to Stata using the Stata Function Interface (sfi), the Python module that we provide.

6. Bayesian predictions, multiple chains, and more

We have lots of new Bayesian features.

We now have multiple chains. Has the MCMC converged? Estimate models using multiple chains, and reported will be the maximum of Gelman-Rubin convergence diagnostic. If it has not yet converged, do more simulations. Still hasn’t converged? Now you can obtain the Gelman-Rubin convergence diagnostic for each parameter. If the same parameter turns up again and again as the culprit, you know where the problem lies.

We now provide Bayesian predictions for outcomes and functions of them. Bayesian predictions are calculated from the simulations that were run to fit your model, so there are a lot of them. The predictions will be saved in a separate dataset. Once you have the predictions, we provide commands so that you can graph summaries of them and perform hypothesis testing. And you can use them to obtain posterior predictive p-values to check the fit of your model.

There’s more. Click the title.

7. Extended regression models (ERMs) for panel data

ERMs fits models with problems. These problems can be any combination of (1) endogenous and exogenous sample selection, (2) endogenous covariates, also known as unobserved confounders, and (3) nonrandom treatment assignment.

What’s new is that ERMs can now be used to fit models with panel (2-level) data. Random effects are added to each equation. Correlations between the random effects are reported. You can test them, jointly or singly. And you can suppress them, jointly or singly.

Ermistatas got a fourth antenna.

8. Importing of SAS and SPSS datasets

New command import sas imports .sas7bdat data files and .sas7bcat value-label files.

New command import spss imports IBM SPSS version 16 or higher .sav and .zsav files.

I recommend using them from their dialog boxes. You can preview the data and select the variables and observations you want to import.

9. Flexible nonparametric series regression

New command npregress series fits models like

y = g(x1, x2, x3) + ε

No functional-form restrictions are placed on g(), but you can impose separability restrictions. The new command can fit

y = g1(x1) + g2(x2, x3) + ε

y = g1(x1, x2) + g3(x3) + ε

y = g1(x1, x3) + g2(x2) + ε

and even fit

y = b1x1 + g2(x2, x3) + ε

y = b1x1 + b2x2 + g3(x3) + ε

I mentioned that lasso can perform inference in models like

. dsregress y x1, controls(x2-x999)

If you know that variables x12, x19, and x122 appear in the model, but do not know the functional form, you could use npregress series to obtain inference. The command

. npregress series y x12 x19 x122, asis(x1)

fits

y = b1x1 + g2(x12, x19, x122) + ε

and, among other statistics, reports the coefficient and standard error of b1.

10. Multiple datasets in memory, meaning frames

I’m a sucker for data management commands. Even so, I do not think I’m exaggerating when I say that frames will change the way you work. If you are not interested, bear with me. I think I can change your mind.

You can have multiple datasets in memory. Each is stored in a named frame. At any instant, one of the frames is the current frame. Most Stata commands operate on the data in the current frame. It’s the commands that work across frames that will change the way you work, but before you can use them, you have to learn how to use frames. So here’s a bit of me using frames:

. use persons

. frame create counties

. frame counties: use counties

. tabulate cntyid

. frame counties: tabulate cntyid

Well, I’m thinking at this point, it appears I could merge persons.dta with counties.dta, except I’m not thinking about merging them. I’m thinking about linking them.

. frlink m:1 cntyid, frame(counties)

Linking is frame’s equivalent of merge. It does not change either dataset except to add one variable to the data in the current frame. New variable counties is created in this case. If I were to drop the variable, I would eliminate the link, but I’m not going to do that. I’m curious whether the counties in which people reside in persons.dta were all found in counties.dta. I can find out by typing

. count if counties==.

If 1,000 were reported, I would now drop counties, and it would be as if I had never linked the two frames.

Let’s assume count reported 0. Or 4, which is a small enough number that I don’t care for this demonstration. Now watch this:

. generate relinc = income / frget(counties, medinc)

I just calculated each person’s income relative to the median income in the county in which he or she resides, and median income was in the counties dataset, not the persons dataset!

Next, I will copy to the current frame all the variables in counties that start with pop. The command that does this, frget, will use the link and copy the appropriate observations.

. frget pop*, from(counties)

. describe pop*

. generate ln_pop18plus = ln(pop18plus)

. generate ln_income = ln(income)

. correlate ln_income ln_pop18plus

I hope I have convinced you that frames are of interest. If not, this is only one of the five ways frames will change how you work with Stata. Maybe one of the other four ways will convince you. Visit the overview of frames page at stata.com.

11. Sample-size analysis for confidence intervals

The goal is to optimally allocate study resources when CIs are to be used for inference or, said differently, to estimate the sample size required to achieve the desired precision of a CI in a planned study. One mean, two independent means, or two paired means. Or one variance.

12. Nonlinear DSGE models

DSGE stands for Dynamic Stochastic General Equilibrium. Stata previously fit linear DSGEs. Now it can fit nonlinear ones too.

I know this either interests you or does not, and if it does not, there will be no changing your mind. It interests me, and what makes the new feature spectacular is how easy models are to specify and how readable the code is afterwards. You could almost teach from it. If this interests you, click through.

13. Multiple-group IRT

IRT (Item Response Theory) is about the relationship between latent traits and the instruments designed to measure them. An IRT analysis might be about scholastic ability (the latent trait) and a college admission test (the instrument).

Stata 16’s new IRT features produce results for data containing different groups of people. Do instruments measure latent traits in the same way for different populations?

Here is an example. Do students in urban and rural schools perform differently on a test intended to measure mathematical ability? Using Stata 16, you can fit a 2-parameter logistic model comparing the groups by typing

. irt 2pl item1-item10, group(urbanrural)

What’s new is the group() option.

Does an instrument measuring depression perform the same today as it did five years ago? You can fit a graded-response model that compares the groups by typing

. irt grm item-item10, group(timecategory)

And IRT’s postestimation graphs have been updated to reveal the differences among groups when a group() model has been fit.

The examples I mentioned both concerned two groups, but IRT can handle any number of them.

14. Panel-data Heckman-selection models

Heckman selection adjusts for bias when some outcomes are missing not at random.

The classic example is economists’ modeling of wages. Wages are observed only for those who work, and whether you work is unlikely to be random. Think about it. Should I work or go to school? Should I work or live off my meager savings? Should I work or retire? Few people would be willing to make those decisions by flipping a coin.

If you worry about such problems and are using panel data, the new xtheckman command is the solution.

15-21. Seven more new features

I will summarize the last seven features briefly. My briefness makes them no less important, especially if they interest you.

15. NLMEs with lags: multiple-dose pharmacokinetic models and more can now be fit by Stata’s menl command for fitting nonlinear mixed-effects regression. This includes fitting multiple-dose models.

16. Heteroskedastic ordered probit joins the ordered probit models that Stata already could fit.

17. Graph sizes in inches, centimeters, and printer points can now be specified. Specify 1in, 1.4cm, or 12pt.

18. Programmers: Mata’s new Quadrature class numerically integrates y = f(x) over the interval a to b, where a may be -∞ or finite and b may be finite or +∞.

19. Programmers: Mata’s new Linear programming class solves linear programs using an interior-point method. It minimizes or maximizes a linear objective function subject to linear constraints (equality and inequality) and boundary conditions.

20. Do-file Editor: Autocompletion and more. The editor now provides syntax highlighting for Python and Markdown. And it autocompletes Stata commands, quotes, parentheses, braces, and brackets. Last but not least, spaces as well as tabs can be used for indentation.

21. Stata for Mac: Dark Mode and tabbed windows. Dark mode is a color scheme that darkens background windows and controls so that they do not cause eye strain or distract from what you are working on. Stata now supports it. Meanwhile, tabbed windows conserve screen real estate. Stata has lots of windows. With the exception of the Results window, they come and go as they are needed. Now you can combine all or some into one. Click the tab, change the window.

 

That’s it

The highlights are 58% of what’s new in Stata 16, measured by the number of text lines required to describe them. Here is a sampling of what else is new.

  • ranksum has new option exact to specify that exact p-values be computed for the Wilcoxon rank-sum test.
  • New setting set iterlog controls whether estimation commands display iteration logs.
  • menl has new option lrtest that reports a likelihood-ratio test comparing the nonlinear mixed-effects model with the model fit by ordinary nonlinear regression.
  • The bayes: prefix command now supports the new hetoprobit command so that you can fit Bayesian heteroskedastic ordered probits.
  • The svy: prefix works with more estimation commands, namely, existing command hetoprobit and new commands cmmixlogit and cmxtmixlogit.
  • New command export sasxport8 exports datasets to SAS XPORT Version 8 Transport format.
  • New command splitsample splits data into random samples. It can create simple random samples, clustered samples, and balanced random samples. Balance splitting can be used for matched-treatment assignment.

I could go on. Type help whatsnew15to16 when you get your copy of Stata 16 to find out all that’s new.

I hope you enjoy Stata 16.