I am constantly both amazed and delighted by the commands Stata has that I do not know about yet. I have bought over 20 books from Stata Press and others on Stata and Mata and continue to learn both languages. [...] I am very impressed with Mata’s concise and powerful structure as related to vectors and matrices. What sets you apart beyond using my past resources of R and/or MATLAB is the excellent and up-to-date documentation and webinars for your end users, as well as logic put into both Mata and Stata functions, commands, structures, etc.
At least for my current project, Stata 17 is much faster than Stata
16, which I presume is due to the Intel MKL. I am simulating a small
VAR model and obtaining bootstrap forecast intervals within each
round of simulation, and all of the code is written in Mata.
Stata 17 is using one core, and its output log reads
Code: Sim 5 Time 12.97 Cumulative 64.37 ETA 2021-04-28 00:29:15 AM Sim 10 Time 13.35 Cumulative 129.70 ETA 2021-04-28 01:29:40 AM Sim 15 Time 13.05 Cumulative 196.05 ETA 2021-04-28 02:32:10 AMI set up Stata 16 to run the exact same do-file, Stata is using one (different) core, and its output log reads
Code: Sim 5 Time 20.82 Cumulative 103.00 ETA 2021-05-01 08:58:19 AM Sim 10 Time 20.68 Cumulative 208.22 ETA 2021-05-01 11:16:51 AM Sim 15 Time 20.80 Cumulative 312.18 ETA 2021-05-01 11:10:57 AMSince many estimators are written in Mata, I suspect they will run faster as well. Note that I am not working with a particularly large dataset or matrices -- just average sized matrices typical of time-series analysis.
Brian Poi, Principal, Poi Consulting LLC
I just got my Stata 17 and I see huge improvements. I am using AMD Ryzen 7 4800H with 40GbRam. On my Stata 16.1 MP8, my data with 44.7 Million obs used to take 30 seconds to sort; now on stata 17 MP8, it is taking 16.1 seconds to sort. I also ran [a] few other commands where I manipulate the data. Stata 16.1 used to take 8 minutes to complete that task; now Stata 17 takes 3 minutes to complete similar tasks. I am not sure how Intel will perform with the new "Intel Math Kernel Library (MKL)" update. However, I am very happy with the new update.
Ahmed Khan, PhD scholar, University of Waikato School of Accounting, Finance, and Economics, University of Waikato, New Zealand
Among all common statistical packages, Stata is the best for students with very different programming backgrounds: Stata is very easy to pick up for students with no programming experience at all; at the same time, for students with extensive programming experience, Stata provides sufficient flexibility for them to develop their own codes and ado-packages.
Ling Cen, Associate Professor of Finance, The Chinese University of Hong Kong
I am very satisfied with Stata. I was required to use SPSS for my PhD program and decided to survey the market. During a clinical trial design course, one of the professors from Harvard mentioned Stata. Once I tried Stata, I was hooked. The ability to write code and generate publishable quality graphs did it for me.
Scott Howell, PhD, Tudor Bompa Institute, Trident University
Part of what I like about Stata is that it's quasi-"open source"—that is, that econ grad students, stats professors, and etc., can create their own functions and commands and make the ado-files available to the broader Stata community—and that the more popular commands then tend to get included in the next Stata release. [...] SPSS and SAS don't do this.
Dr. Travis Anderson-Bond, Principal Business Analyst, Australia
I really enjoy using Stata and have chosen it as my primary software because the coding is so intuitive, the documentation files are so helpful, and there is always the point and click option available when necessary. Thanks for all you and your team do to produce such a quality resource!
Julie Summey, PhD Student, Graduate Research Assistant, Clemson University, Department of Public Health Sciences
[...] I would like to take the time and effort to thank Stata developers for creating such an extraordinary software to help me through the hardships of econometrics, econ statistics, and econ theory. It is such a great software, and I love to use it. Although I have not been able to use it to its full potential yet, my instructor showed me books written on Stata commands, which make running regressions and statistical analysis a lot faster and simpler.
Adrian Solache, Student, Loyola Marymount University
I'm really impressed with Stata. Today, I went from having zero experience with the package to being able to read in fixed column files, creating and recoding new variables, and a number of other data management tasks. [...] The syntax is so logical and the GUI is very user-friendly.
Linda A. Landon, PhD, ELS, Research Consultant, Research Communiqué
As an assessment coordinator and survey designer I am very pleased with the fact that you added IRT to Stata. I used SPSS for a long while and then about two years ago moved to Stata. I really appreciate the added capabilities and flexibility of Stata.
David I. Hanauer, Professor of Applied Linguistics, Indiana University of Pennsylvania
The July 2016 edition of the Stata Journal is an excellent example of how innovative extensions to Stata are also taking place in the user community. By designing a development platform that can innovate internally, but at the same time, engage thousands in the user community to enhance the functionality, is a winning strategy that distinguishes Stata from other computing platforms.
Murtaza Haider, Associate Professor of Real Estate Management, Ryerson University
For an epidemiologist such as myself, who is trying to learn more about, and I hope soon apply, the [Bayesian] methods, Stata is a "godsend" and removes the barriers that stopped at least me in going further with Bayesian models.
Babak Khoshnood, MD, PhD, Senior Researcher, Institut National de la Santé et de la Recherche Médicale (INSERM)
Just wanted to let you know I have been using stata for a couple weeks now and love it. Wish I had ordered sooner! Will be letting my school know, since they have us using SPSS. The SEM builder is awesome.
Benjamin Powers, Doctoral Candidate, Grenoble Ecole de Management
The Stata Journal and the Stata program package itself comply with the highest quality in statistics and they reflect the best practices in many ways. I also work with SAS EG 7.1 WinBUGS and R according to the needs and have therefore some ability to make a comparison. I am enthusiastic about Stata (14) and the accompanying excellent books and papers.
Laszlo D. Molnar, MD, PhD, Central Administration of National Pension Insurance, Budapest, Hungary
I've been a Stata user since the early-2000s and am impressed with how the product has evolved over the years while remaining so intuitive to use. Stata is clearly in tune with its users. I've not yet been faced with an analytical task that Stata couldn't deliver on.
Neville Arjani, Lead, Payments, Payments Canada
I like how Stata has an easy-to-use "click to select" menu (though code is faster) but it also is comparable to R in having user code that you can instantly get along with help files. It's like the best of both worlds.
Heidi P. Celebi, Doctoral Candidate, Robert H. Smith School of Business, University of Maryland
Every year or two, I have been training 50 to 100 people at my company in using Stata and otherwise getting users of other packages to use it. What I really like ... is that it is not only easy to learn and use at a basic level, but also has tremendous power too. People here rarely have to move outside Stata once they have adopted it.
Dan Sherman, PhD, Managing Economist, American Institutes for Research, Washington, DC
[Stata] has the best of the open-source world and proprietary software. SSC is like an open-source community and you get all of the help and support from Stata.
M.P. "Trey" Marchbanks III, Associate Research Scientist, Texas A&M University
Stata 14 is such a fine statistical software package. The best on the market, to be sure.
Joseph M. Hilbe, Adjunct Professor of Statistics, Arizona State University and Emeritus Professor, University of Hawaii
I am loving [Stata], as there are much more options than SPSS, yet more user friendly than R.
Donny Williams, Sonoma State University
There is a lot to like about Stata, but for an epidemiologist the ease of use of the svy commands is not matched in any other package.
George Savva, School of Health Sciences, University of East Anglia
I have received and downloaded Stata. This is far superior to SPSS and I find it easier to use than SAS.
Audie Daniel Wood, Masters student, Sociology, Idaho State University
Stata's documentation is *amazingly good*.
SPSS's documentation is often inscrutable; Stata's documentation is easy to read, with clear examples and readily understandable formulae and methods, complete with relevant citations. [...] As an educator, I know how hard it is to write about stats clearly and understandably, and I very much appreciate the work done by Stata in this regard.
Stata is the fundamental tool of my job, and it's made even better by knowing that the folks who make it put so much care into not just the software, but all the tools we need to make the software solve the problems we have.
Troy C. Payne, Ph.D., University of Alaska Anchorage
The ability to uninstall a copy, and reinstall on another computer is MAGIC! It has allowed me to have one copy on my PC and another on my Mac. So Stata stays with me as I upgrade computers, and is with me no matter which machine I am using. I love it!
I recently bought Stata in preparation for my PhD comps and it was the best decision I've made in a while. I'd used SAS before (with substantial frustration) and have been flying through the training modules and a book by Alan Acock that I purchased from Stata.
Linda Leitz, CFP, EA, and Co-owner, It's Not Just Money, Inc
When I got my master's in mathematics at MIT in '82 I wrote my own Fortran routines to do gamma distribution regression because there was nothing out there - SAS, SPSS, or the old BMDP that could do it. Now I wouldn't use anything but Stata period. I don't think anything now comes close to the reliability, accuracy, and ease of use of Stata.
Raymond R. Hyatt, Ph.D., President, Freeport Development Strategies, LLC
The new generations of sociologists who are exposed to Stata in our classes love it. It is definitely the leading package in our discipline. I was one of the first adopters in Sociology back when Stata was in version 1.0 (about 1984-5). Stata was a much smaller company back then, but the customer service and support are just as friendly and amazingly responsive today as back then.
Mark Fossett, Professor of Sociology, Texas A&M University
Using Stata for our analyses makes remedial corrections easy: just make a correction to the do-file, and re-run. Clean. Graceful.
Dr. Travis Anderson-Bond, Business Analyst, Australia
Without doubt Stata is an amazing, sophisticated and wonderful statistical package. I am very pleased.
Willy Rice, School of Law, St. Mary’s University
Stata is a very powerful, very reliable, and very flexible program.
Marc Feldesman, Department of Anthropology, Portland State University
These are without competition the best-written and organized manuals for complex software I have ever seen, and certainly the only ones that on occasion make me laugh out loud.
Charles Murray, Bradley Fellow, American Enterprise Institute
I just wanted to let you know that I think your licensing procedure that allows users to install a copy at work and at home is simply wonderful. It allows me to be more productive. Thanks for being so reasonable in an age when reasonableness seems to be lacking!
Jack Laverty, Columbia Gas of Ohio
Stata is awesome! I have been using it to do analyses on panel data. Love it! [...] As a statistician I get to learn new software it seems almost monthly. I have to tell you...Great software, great service too. You guys run rings around SPSS they want to nickel and dime me for everything.
Elaine Bellucci, Statistician, Bellucci Company
Stata = best general purpose statistical program on the market!
Martin Biewen, School of Business & Economics, University of Tuebingen, Tuebingen, Germany
As a biostatistician with responsibility for dealing with daily inquiries from the medical and dental schools in my university, I find that Stata provides the broadest statistical base for data analysis—particularly with regard to the less commonly found routines, such as survival analysis (including Cox’s regression), conditional logistic regression, ordered and multinomial logistic regression, Poisson regression, and negative binomial regression.
Alan Kelly, Department of Statistics, Department Community Health & General Practice, Trinity College, Dublin, Ireland
Stata is easy to use, very powerful, includes most non-linear models people usually think of (and that other packages only include as add-ons).
Jose Camoes Silva, M.I.T. Sloan School of Management
First of all, Stata deserves high marks for being easy to learn. Programming is so easy that I practically never write repetitive commands anymore.
I’ve been very pleased with Stata’s graphing abilities, and we use Stata-generated graphs extensively in our political coverage to explain poll averages and other quantitative data in ways that our competitors can't. Stata performs so well that I'm delighted to credit it at the bottom of every graph that we create.
Stata’s data-handling abilities are outstanding; I use Stata to comb through databases when I’m doing preliminary research before starting on articles, and I’ve developed lots of article ideas from data that Stata makes it easy to examine.
In short, excellent software-the best program of any sort that I’ve ever used. I look forward to opening it up every day.
Jonathan Bruner, Forbes Magazine
Incidentally, one of the really neat features of Stata is the indexing of the STB through the search command. It is a major reason for submitting to the STB rather than to SUGI. One of the things that I like about Stata is that it has the feel of a small company that still cares about its users. This is greatly appreciated.
William Dupont, Vanderbilt University
Off the top of my head here are some reasons why I use Stata almost exclusively:
It has the best coverage for biostatistics/epidemiology as far as I can see.
It is easy to install, but if you have any problems whatsoever, be it with installation or otherwise, they have a hotline (1-800-STATA-PC) which is high quality. They are honest; if they do not know the answer immediately they will admit to it and get back to you on it, rather than try to con you out of the question.
The capabilities include all the standard things (data filtering, statistical models, great graphics) and some non-standard, but useful things like survival (cox, logrank,...), Mantel–Haenszel, logistic (probit, tobit,...), etc. Plus, if what you want is not there, they have a programming language built in.
Works on all sorts of platforms, including the big three: Unix, Windows, and Macintosh.
Good on-line help.
Good manuals (maybe a smidgen difficult to use, but they make up for it in thoroughness).
Forever updating. For example, the new version for Windows has a spreadsheet data editing feature that is really snazzy.
Interfacing with printers, word processors, etc., is good.
They have an ongoing bulletin with useful additions, articles, etc. (I am an associate editor on their bulletin. No remuneration. Above is honest opinion.)
All this is off the top of my head, and I am sure I have left some important stuff out. Contact them directly, they take pride in their work and they have a good product. I am sure they can give you a better pitch than I have.
Marcello Pagano, Harvard School of Public Health
I’ve used a lot of stat packages over the years, but I find that I’m using Stata 95% of the time now. It’s wonderful! Its speed and power are much touted, but its simplicity for beginners is perhaps one of its best features.
Rodney Hayward, University of Michigan’s Schools of Medicine & Public Health, Ann Arbor VA’s Center for Clinical Management Research
Most impressive is the fact that you guys find fixes for every problem or question that we have, quickly and efficiently. And little by little, as Stata develops, we are dropping the other programs we use, too. I don’t know of any other software company that responds as well to its users’ questions, suggestions, and comments.
Julien Teitler, Department of Sociology, University of Pennsylvania
With software, one is often faced with the choice of a program that is easy to learn (but limited) or one that is hard to learn and use (but powerful). Stata is both easy to learn and also very powerful. Stata is easy to learn in two ways. First, it offers a point-and-click interface that you can use if you do not know the name of a command, or the particulars of using it. Second, it also offers commands that have a simple and consistent structure, making them easy to learn. Consider the command to run a regression predicting "y" from "x1 x2 x3". You type "regress y x1 x2 x3". Say you want to run this as a robust regression; then type "rreg y x1 x2 x3". To run this as a logistic regression, just type "logistic y x1 x2 x3". If I told you that the command for ordinal logistic regression was called "ologit", you would probably guess that you would use it by typing "ologit y x1 x2 x3", and you would be right.
Stata is also very powerful. Stata has both a built-in programming language as well as a matrix programming language. In fact, most Stata commands are written in these languages, and you can view the source code for any of these commands yourself. You could make your own copy of any of these Stata commands, customizing it as you wish. You can also use these built-in programming tools to create your own Stata programs. This is why I say that Stata is easy to learn, but difficult to outgrow.
Michael Mitchell, senior statistician at the USC Children's Data Network, author of four Stata Press books, and former UCLA statistical consultant who envisioned and designed the UCLA Statistical Consulting Resources website