Note: This article was originally published on Mar 27th, 2014 and updated on Sept 12th, 2017
We love comparisons!
From Samsung vs. Apple vs. HTC in smartphones; iOS vs. Android vs. Windows in mobile OS to comparing candidates for upcoming elections or selecting captain for the world cup team, comparisons and discussions enrich us in our life. If you love discussions, all you need to do is pop up a relevant question in middle of a passionate community and then watch it explode! The beauty of the process is that everyone in the room walks away as a more knowledgeable person.
I am sparking something similar here. SAS vs. R has probably been the biggest debate the data science industry might have witnessed. Python is one of the fastest growing languages now and has come a long way since it’s inception. The reason for me to start this discussion is not to watch it explode (that would be fun as well though). I know that we all will benefit from the discussion.
This has also been one of the most commonly asked questions to me on this blog. So, I thought I’ll discuss it with all my readers and visitors!
Probably yes! But I still feel the need for discussion for following reasons:
So, without any further delay, let the combat begin!
Here is a brief description about the 3 ecosystems:
I’ll compare these languages on following attributes:
I am comparing these from point of view of an analyst. So, if you are looking for purchasing a tool for your company, you may not get complete answer here. The information below will still be useful. For each attribute I give a score to each of these 3 languages (1 – Low; 5 – High).
The weightage for these parameters will vary depending on what point of career you are in and your ambitions.
SAS is a commercial software. It is expensive and still beyond reach for most of the professionals (in individual capacity). However, it holds the highest market share in Private Organizations. So, until and unless you are in an Organization which has invested in SAS, it might be difficult to access one. Although, SAS has brought in a University Edition that is free to access but it has some limitations. You can also use Jupyter Notebooks in there!
R & Python, on the other hand are completely free. Here are my scores on this parameter:
SAS – 3
R – 5
Python – 5
SAS is easy to learn and provides easy option (PROC SQL) for people who already know SQL. Even otherwise, it has a good stable GUI interface in its repository. In terms of resources, there are tutorials available on websites of various university and SAS has a comprehensive documentation. There are certifications from SAS training institutes, but they again come at a cost.
R has the steepest learning curve among the 3 languages listed here. It requires you to learn and understand coding. R is a low level programming language and hence simple procedures can take longer codes.
Python is known for its simplicity in programming world. This remains true for data analysis as well. While there are no widespread GUI interfaces as of now, I am hoping Python notebooks will become more and more mainstream. They provide awesome features for documentation and sharing.
SAS – 4.5
R – 2.5
Python – 3.5
This used to be an advantage for SAS till some time back. R computes every thing in memory (RAM) and hence the computations were limited by the amount of RAM on 32 bit machines. This is no longer the case. All three languages have good data handling capabilities and options for parallel computations. This I feel is no longer a big differentiation. They’ve all also brought on Hadoop and Spark integrations, with them also supporting Cloudera and Apache Pig.
SAS – 4
R – 4
Python – 4
SAS has decent functional graphical capabilities. However, it is just functional. Any customization on plots are difficult and requires you to understand intricacies of SAS Graph package.
R has highly advanced graphical capabilities along with Python. There are numerous packages which provide you advanced graphical capabilities.
With the introduction of Plotly in both the languages now and with Python having Seaborn, making custom plots has never been easier.
SAS – 3
R – 4.5
Python – 4.5
All 3 ecosystems have all the basic and most needed functions available. This feature only matters if you are working on latest technologies and algorithms.
Due to their open nature, R & Python get latest features quickly. SAS, on the other hand updates its capabilities in new version roll-outs. Since R has been used widely in academics in past, development of new techniques is fast.
Having said this, SAS releases updates in controlled environment, hence they are well tested. R & Python on the other hand, have open contribution and there are chances of errors in latest developments.
SAS – 4
R – 4.5
Python – 4.5
Globally, SAS is still the market leader in available corporate jobs. Most of the big organizations still work on SAS. R / Python, on the other hand are better options for start-ups and companies looking for cost efficiency. Also, number of jobs on R / Python have been reported to increase over last few years. Here is a trend widely published on internet, which shows the trend for R and SAS jobs. Python jobs for data analysis will have similar or higher trend as R jobs:
The graph below shows R in Blue and SAS in Orange.
Overall, the market based on languages can be pictured as such:
SAS – 4
R – 4.5
Python – 4.5
R and Python have the biggest online communities but no customer service support. So if you have trouble, you are on your own. You will get a lot of help though.
SAS on the other hand has dedicated customer service along with the community. So, if you have problems in installation or any other technical challenges, you can reach out to them.
SAS – 4
R – 3.5
Python – 3.5
Deep Learning in SAS is still in it’s beginning phase and there’s a lot to work on it.
On the other hand, Python has had great advancements in the field and has numerous packages like Tensorflow and Keras.
R has recently added support for those packages, along with some basic ones too. The kerasR and keras packages in R act as an interface to the original Python package, Keras.
SAS – 2
Python – 4.5
R – 3
Following are some more points worthy to note:
We see the market slightly bending towards Python in today’s scenario. It will be pre-mature to place bets on what will prevail, given the dynamic nature of industry. Depending on your circumstances (career stage, financials etc.) you can add your own weights and come up with what might be suitable for you. Here are a few specific scenarios:
Strategically, corporate setups that require more hands-on assistance and training choose SAS as an option.
Researchers and statisticians choose R as an alternative because it helps in heavy calculations. As they say, R was meant to get the job done and not to ease your computer.
Python has been the obvious choice for startups today due to its lightweight nature and growing community. It is the best choice for deep learning as well.
Here is the final scorecard:
These are my views on this comparison. Now, it’s your turn to share your views through the comments below.
I can say by experience that R is a lot more fun than SAS. Exploring all the packages and conversations on Stackoverflow . . .
I felt python scikit has better documentation compared to R. Also for handling unstructured data python seems better. Any other views if you have please share.
Thanks for this.
I agree
Absolutely True! R is simpler to learn than SAS in my view.
Good one Kunal! Liked it.
I liked each and every article of yours..................... thanks a lot......
Sir, If MATLAB would be included in comparison then it would be best because MATLAB is also very popular and used by a umpteen number of people for data analytics..so plz include MATLAB also in comparison then it will give us clear picture...
Chandan, Matlab lies somewhere in between on most of the parameters. I had used Matlab long time back and have not followed the software very actively. Hence some of these views may not be entirely accurate / updated. If you find this to be the case, please update me. Here are my views: Availability / Cost - Matlab is expensive, though lower than SAS. Exact quote might vary depending on the requirements. Ease of learning - I found SAS easier to learn and would rate MATLAB to be easier than R Data handling capabilities - At par with other tools. Graphical capabilities - Good, similar to R Advancements in tool - Similar to SAS. Happens only in releases Job scenario - Matlab has lower market share as compared to SAS and R. It would be higher than Python as of now. Customer service support and Community - Good Hope this helps. Regards, Kunal
Good ,entire blog gives clear insight of these three tools .
Good !! Entire blog gives clear insight of these 3 tools .
Very Well written and balanced article Kunal. Being bit biased to SAS, I'd recommend students and seasoned Analytic professionals to look at SAS Visual Analytics, and SAS Visual Statistics, the latest offerings from SAS which are changing the way people see Commodity Analytics solutions.
Hello guys, Thanks for starting this topic. In my opinion languages of the future for analytics are as follows: R => No. 1 => King (Currently R is the King but in future Python will give tough fight to R as Python is both General purpose programming language and data analysis tool due to enhanced libraries like Pandas, Scipy, Numpy as opposed to R which is only statistical analysis tool......In data science domain 70% time of Data scientist 's job is data munging that is cleaning the data and only 30% is real statistical analysis, hence Python seems much more robust and 4 years down the line Python will be the King and R will be the Queen) Python => No. 2 => Queen I don't see any future for SAS.....I think in next 4 years, SAS will not be in the market, it will be completely dead. These are not predictions ....these are real fates of these languages in future.
Dear King, While you are right on high adaption on R and Python bringing good capabilities to the table. It is too early to write-off / discount SAS. Most of the organizations which are in SAS would continue to use it. Further, there are some interesting products SAS has offered lately - Visual Analytics, High Performance Computing. Also, if cost is the biggest reason for SAS to lose market, they can always play around with price points. Thanks, Kunal
One big factor that will keep SAS in the market is government regulation. Banks, clinical research organizations and any other industry that is required to report to the government will be required to use SAS.
Adding few important points, comparing SAS vs R is not comparing apple to apple. If we compare R with "Complete SAS" then R is nowhere in the league. If we compare R with Base SAS, then R is better, as it gives all the facility for free which base SAS gives. R is hard core programming, but open source (free), which can be used to develop enterprise wide analytics application. It has thousands of built in function to solve complex analytics problem. For non-programming background, little difficult to expertise. R is widely used in academic and SME. SAS is a complete end to end solution from Data management to Data visualization to ETL to BI report to Advance analytics. Even with SAS EIS and SAS EF we can develop enterprise wide large scale application. Also SAS SCL can be also be used as general purpose OO programming language like Java, C++. More important aspect, SAS EG or E-miner has user friendly GUI, which can be used by non-programmer comfortably. In industry SAS is widely used due to below few reasons. 1. Data security is very high, for this reason in BFSI domain SAS is no. 1 choice. 2. SAS has over 250 industry specific “point and click” solution like credit and market risk, Asset/ Performance management, Fraud/ Pricing/ Marketing analytics. 3. SAS DI is a very powerful ETL tool which augments the power of SAS in Data management. 4. SAS BI provides the facility to produce world class dashboard in real time. 5. In corporate time has more value than product cost. 90% of Fortune 100 companies uses SAS. 6. SAS visual analytics analyzes Big data in real time with great visualization as “Tableu”. In future when completion grows, Base SAS can be made free as SAS University edition. SAS is much more than just an analytic tool and it will always remain dominant in future.
Great post Kunal, thank you!
Nice Post ! In present Scenario people are migrating towards Open-Source platforms rather than counting days with proprietary Software products. Well, if we consider job market and usage only in India "SAS" grabs the kings place. When we look globally "R" & "Python" are leading all their way in analytics Industry. People mostly tend towards R and python due to their drastic raise in development and documentation in terms of Research and Education. As of now there might be some market for SAS, but in future it's market share and usage might decline when compared to it's closest rivals.
Hi All, It's a really nice post and very-very healthy discussion... In my opinion also future of SAS is not as easy as it's present. R is growing very fast and because it is being accepted and widely used in Universities from where the new breed of the people will come they will be well equipped with R and/or Python. But since SAS is quite widely used in Financial Industries where correctness, stability, accuracy, adaptability in market i.e. acceptance in market, reliability and high support environment is required which is still quite high with SAS will not allow SAS to go out from Financial Sector (from other sector it may go early), it will be replaced in future but not in just 4-5 years, later on when the market share of SAS will start falling, SAS may come up with different pricing strategy, offers which might look lucrative to financial industries at that time. Hence learn well what is widely used and accepted in current market and start learning the language of future, definitely learning 2-3 languages is always good than to stick to a single language. Keep exploring new advancements in these languages and keep discussing always. Thanks Kunal for such a nice and healthy topic to bring on table for discussion and thanks to all contributors. Please share some more information/insight on this topic. Regards Kumar Ravi
Thanks Kumar Ravi! Well articulated.
Very helpful article and very clearly structured. Congratulations! I have had experience with all 3 languages/programs, and I think your analysis is quite valid; albeit with an Indian perspective. I would like to add that I find SAS JMP a major addition to SAS - I believe it is free with base SAS, and otherwise it is a very cheap alternative as a stand-alone. It's forte is preliminary data analysis, but with its own programming language (JSL, which is somewhat C-like) it can be used to automate to almost any degree). It has the shortest, smoothest learning curve by a country mile! Also its partial integration with EXCEL is extremely useful with clients with low levels of sophistication. These considerations mean that, working in an educational (50 PhD students, but non DGI), I am using this tool almost exclusively, to the detriment of my facility with the other 3 main languages.
Thanks Roxy! Nice point about JMP. A point worth mentioning is the Excel add-on for SAS. I have used it frequently and found it very useful way to import data in SAS and then use Pivot tables over and above it. Regards, Kunal
really its a great discussion..
Hi Kunal , Very informative thread on SAS & R
vry informative..
Thank you sir... I have read few of your blogs.. plz keep writing and guiding us.. I'm a student from NIIT Chennai, pursuing Business Analytics course.
hi shakthi I would like to know how the business analytics course in NIIT in Chennai and which branch you are taking up the course. I'm planning to take up the course but still in search for a right institute.So pls help me with some info about the traning and the placement oppurtunities that they provide for business analytics. Here is my email id: [email protected] Thank u Vasavi
R is the BJP(if they win), Python( like AAP) , for SAS (CONGRESS?)
sir, should i go for SAS TRAINING & CERTIFICATION or BUSINESS analytics training using SAS,EXCEL. WHICH WILL GIVE BETTER JOB prospective for FRESHERS????? PLSS suggest some institutes in bangalore Either online or class room I have completed B tech electronics and have worked in a core company whose experience wont count at all I have also appeared final yr MBA this yr. PLSS reply...
Pranay, The training depends on how you want to pursue your career. If you want to get into technical data analyst role - SAS training / certification is more useful. However, if you want to build a career as business analyst, certification courses are better as they provide more case studies, are typically more spaced out so that you can apply concepts at work. Online / offline have their own pros and cons. Online provides flexibility, where as offline provides more opporunity to interact with instructor and class. See what fits you better. Regards, Kunal
Hi, Great article. I enjoyed looking and thinking about how you scored each language on each point. You obviously put real thoughts into this topic. Here's a different (and complementary) perspective from industry. I was just at Big Data Paris earlier this week. Some observations on the trade show floor and the presentations: 1) SAS had a booth, but it was empty most of the time. Not a single presentation mentioned SAS as a data analysis platform to invest in. My interpretation of this is similar to what's been said so far: they have a large installed base, but not where new developments and new business investments are going. Having said that, SAS isn't going away. Remember COBOL, a language of the 1970's. It was still a BIG deal in 2000 with the infamous Y2K problem that all large businesses still running mainframes had. In a similar way, SQL is the leader in databases today, but Hadoop is quickly taking over market share due to the power of map-reduce and its open source aspect, and pushing SQL aside. Perhaps it's no surprise that Intel invested $740M in Cloudera this week (Hadoop integration leader) and canceled their own big data program, pushing Cloudera's valuation to $4.1B (they are still a privately owned startup). Hadoop is the new generation, while SQL is the old guard. Perhaps the same can be said with SAS vs. R/Python? 2) There was a huge focus on Hadoop as the DB platform, coupled with R as the main engine for serious data analytics. Many presentations couple that with several other specialized tools for simple visualizations (Tableau, etc.) so that the business can enable non technical users fairly easy and provide simple ways to explore and visualize data straight out of the Hadoop DB ecosystem. In other words, the simpler tasks of SAS are being replaced by visualization software tools like Tableau while the more serious data analysis seems to be going to R which, as stated elsewhere in this post, is experiencing growing momentum with new packages, etc almost every day. 3) Very little mention of Python as the core data analysis platform. No doubt Python is quickly becoming the language of choice for general purpose work and many data analysis tasks (Pandas, matplotlib, etc.), but at least in this conference, it was not seen as what to use for a core enterprise big data analysis platform. Please don't misinterpret this comment however. Python was still mentioned, but more in light of general purpose work and some analytics. Key takeaway: it's important to realize that today's installed base is not the future. Where investments are going is what the future is about. And at least from this conference, it looks like enterprises are investing heavily into the Hadoop / R ecosystem (and Python). Given these are open source platforms and adopted by leading companies like Google and facebook, you can expect the momentum to accelerate. This does not bode well for SAS's future in my opinion, which will become a legacy product (but won't go away for a long time still). I hope this sheds some light with a (different) industry perspective. Thanks again for the great article.
Jean-Marc, Very well put across and a very clear perspective on how things stand. I have a couple of points to add: 1. The evolution is at a different stage in different markets. In India, for example, SAS Annual forum is still one of the most sought after event to attend. 2. You are talking about a forum on big data, which is still a low percentage (but with high growth) of the entire spectrum. I agree with you that future lies in open source ecosystems, but for people looking to enter industry, SAS may still provide the best hope (for at least couple of years, if not more). Regards, Kunal
@ Jean-Marc. I like the your comment
I really can't agree that R can handle data as well as SAS. How to handle gigabyte size data with R? Using Revolution R, but that comes at a cost, this would negate the cost aspect of R. So SAS is still the way to go if you routinely have to manipulate gigabyte size data
ZJ, This used to be a big problem with R. However, a lot of work has happened on this front on R. You will find some of the work if you research. Following is a good article with some handy tips: http://theodi.org/blog/fig-data-11-tips-how-handle-big-data-r-and-1-bad-pun Thanks, Kunal
Hi Kunal, Very well written, clear and concise article!!! I was searching for such an article since many days, finally found yours!! I am currently working as a Solution Developer in Tata Technologies,Pune with almost 5 years of experience in enhancement and support of legacy IT applications belonging to a major European automotive giant. Recently I have completed a course in Business Analytics where I have learnt Excel VBA, SAS, SPSS and application of analytical concepts in SAS and SPSS. I have some queries : 1) I am interested to work as an Analytics Consultant. So which certification in SAS would be beneficial ?Is BASE SAS certification enough? OR do I need to do 'SAS Certified Statistical Business Analyst Using SAS 9'? 2) Would it be better to enter the industry as a SAS Developer\Programmer and then move into the Analytics profile? 3) The job openings for Analytics in Pune is very less compared to Bengaluru, Noida/Gurgaon and Mumbai.Also many companies are asking for minimum 2-3 yrs of relevant experience. How to get this relevant experience?Also what is your take on the analytics job market in Pune in the near future?
Varun, Here are answers to your questions: 1. SAS Certified Statistical Business Analyst Using SAS 9 is more beneficial. You can also attend Predictive modeling. 2. If you can get a break through directly, that is preferred profile rather than entering as a developer. 3. Pune offers limited job opportunities. Bangalore, Deli NCR and Mumbai combined will have 90%+ market share in jobs. It would help if you are in any of these cities. Consultancies typically hire straight out of college, but because of high demand they are looking at lateral hires. You can look at them (e.g. Fractal, Mu-Sigma etc.) Thanks, Kunal
I would say that R is more and more popular in academies and universities and those people being skilled in R will make R more competitive in market. At the same time, SAS is also investing a lot to compete and its integration with big data with Hadoop ecosystem is accelerating. And as a enterprise analytical platform, SAS still has its advantage especially in Financial sectors which does not take cost as the highest consideration in selecting tools. At the same time, SAS has been more and more cheaper for university students to compete R in future users.
Hi Kunal, I am working as Java Developer with 22 months of experience but want to make a career change . I want to become a data scientist.which certification of SAS i should start with and other things to get a job in this field. - I am doing some courses on Coursera like Machine Learning,Big Data in Education,Statistical Inference
Vipin, Coursera is the right place to start this learning. You can start applying them by participating in competitions on Kaggle. Look at this article for some more handy tips. Thanks, Kunal
Nice Article Kunal. It has given real insights.
Take a look at following project, you might give python higher scores in most of parameters http://numba.pydata.org/ http://blaze.pydata.org/index.html http://bokeh.pydata.org/ And, it's better to mention pandas and statsmodel in python. It learns something from R (such like Dataframe). And it runs much faster then R.
Willy, Thanks for these links. Have just glanced through them and they are definitely representative of the work happening on Python. Pandas have obviously been a significant update to Python and has made it a worthy competitor of R. Thanks, Kunal
Kunal , You article is significant to me ,because i have been searching for such information for a while now. I really want to learn one of this languages,which one should i start with and were?,do you have any online training course on this ?. Please let me know Regards
Dire, As mentioned in the article, the right answer about which one to start actually depends on your experience and role. Let me know these details and I might be of more help. Thanks, Kunal
Kunal, That's a brillint article that has added value to a lot of people. It will enable a lot students graduating to get a perpespective regarding the current industry scenario. This article and the comments that follow makes my task easier in counselling students. Thanks Once again.
Thanks Prof. Toby for your appreciation. This value add to audience is what keeps us ticking. Regards, Kunal
You count off for R because it doesn't have support. That really isn't correct. Revolution Analytics provides professional support for R much the way that SAS (the company) provides support for SAS (the product family). The only differences are that the company doesn't have the same name as the product and in the case of R, much of the code is written by the community.
Ted, Thanks for your comment. If I look at the Revolution Analytics for comparison, the cost advantage I have provided to R goes away. We can compare it either ways. Thanks, Kunal
Hi Kunal, In continuation of my previous post, can you suggest me the training materials/books/onsite websites for 'SAS Certified Statistical Business Analyst Using SAS 9 ' exam ? I have enquired regarding the official SAS crash course training for the same, but the cost is a whopping Rs.72,000/-..... Also, could you give me some tips on mastering statistics concepts and SAS,Excel and SPSS? I guess your answer will be to practise as much as possible right? Also, could you tell me some links where I can get real-life case studies and datasets? I have found some which are given below: 1. http://www.principlesofeconometrics.com/sas.htm 2. http://support.sas.com/learn/statlibrary/statlib_eg4.2/top_data.htm 3. http://support.sas.com/learn/statlibrary/statlib_eg4.2/anova_1.htm#KB5
Varun, If money is a constraint, look at Jigsaw Academy. They have good course and about half of the price sompared to SAS. It is also more spaced out. Will share case studies through an upcoming article. Regards, Kunal
Have you try the Wolfram Language? It does everything for me. Personally, I use the Wolfram Language and Python for all my data science work.
Steeve, Had tried Wolfram language some years back. Have not tried it recently. It would be great if you can add some details on the attributes of comparison. It will help other audience on the blog. Thanks, Kunal
Kunal, Thanks for hosting the discussion. You mentioned that R's in-memory data processing is no longer a limitation. Can you please clarify what are the techniques to overcome this limitation that R has?
Kishon, What I meant was that with 64 bit architecture, there is no absolute limit to what R can handle (which was not the case with 32 bit machines). Regards, Kunal
Great presentation. Very informative, well structured. Take a look also to this, if you do have the time: http://www.chrisstucchio.com/blog/2013/hadoop_hatred.html
Hi Kunal, I have gone through the all comments on this block. It's very informative & well structured. I want to switch the technology. I have 3.5 years in IBM(SPSS) Dimensions(Scripting) field.Currently working on Dimensions-6.0 version in Hyderabad. I want to learn SAS as new technology. Can you let me know, is it good to start with SAS technology with base & advanced sas course? Is their job opportunity in Mumbai or Pune for SAS which allows me to show my current technology experience? Also let me know, which domain I need to select (Clinical\Finance\Insurance) to make career? Regards, Firoj Shaikh.
Firoj, Here are my suggestions: 1. Base and advanced SAS are good place to start 2. There are multiple opportunities in each of the city. You can check out our jobs page for more details. 3. I would say to learn SAS with a broad application and choose the domain later on. Regards, Kunal
Hi Kunal, Thanks a lot for this article!! I was confused to pick the right course to go for. The article has really helped me. could you please help me with what kind of knowledge would one require to take up R? (I'm relatively new to analytics)
Seleena, There are various courses running online - you can check out Jigsaw for a paid / assisted version or Data Science specialization for a free community based learning. None of these course require you to know analytics before hand. Regards, Kunal
Hi, The post and discussion here is very informative. Thanks for this. I would like to take guidance on what path to choose in Analytivs. I have 7 years of experience in BI and data warehousing and want to enter into Analytics. Should I start with SAS or R? Moreover, since I am more interested in consulting what courses should I go for? Thanking in advance. Neha
Neha, SAS would be my recommendation, given its market share in jobs. Also, it would also help in your BI role. Regards, Kunal
Dear Kunal, Thanks for this Post, its just getting more confusing for me though. I am into SAP MM Support & wanna move into Business Analytics. So want to learn course on Analytics, but now dont know from where to start. 1. Kindly provide me info for Classroom foundation courses on Analytics. 2. Will they provide SAS or R software ? 3. Hows EduPristine for learning Business Analytics ? I am thinking on joining it as its classroom. Kindly reply, thanks in advance.
Mayur, We are coming up with a comparison in short time, so stay tuned for recommendation on it. Edupristine has mixed reviews, so I would not recommend it. A better way might be to join SAS certification from one of the certified partners or taking up online instructor lead courses. Regards, Kunal
Good informative article mr kunal,i am learning my masters in data science and we are taught data aanalysis using spss ,how spss is different from sas and other programming paltform and languages,let me know.kindly reply . thanks
Vittal, SPSS falls in the same category as SAS - a proprietary software. Your focus should be on the fundamentals of various techniques, rather than the tool. If you pick up SPSS, you can always learn SAS. Regards, Kunal
Thanks very much for your thoughts, Kunal. We are about to open a financial instruments trading company. We will trade currencies, stocks and others. We originally are manual traders who want to "translate" our efforts into alogs. You know this needs programming , stats and back testing. We will start from Dubai and we are totally ignorant of what tools we shall have. Do you have an advice in that? I mean: 1. What kind of "technical" person we have to put as an architect to manage all this; 2. What kind of programming tools a programmer employed with us must have; 3. What kind of programs such as SAS, R, Python or Matlab we shall have "we can pay for expensive ones but we are only keen to have THE BEST"; 4. We are very keen to kill latency, so also what kind of persons and tools we shall have for that. I am searching and reading and have a lot of info that indeed does not give me the STRONG and VALID start up. If you could help. "I don't know if we can arrange something business wise". Thanks a lot in advance. Fadi
Fadi, It is best to take this discussion one on one, as I need to understand the background in more details before advising any thing. Please drop me a mail on [email protected] and we can take the discussion forward. Regards, Kunal
SAS is NOT AT ALL easier to learn compared to R. The very old fashioned macro language make SAS really awkward. You can easily forget everything and code like a novice in SAS if you don't code it constantly. Why? Because there is no consistent logic and intuition in SAS. Coding in SAS, you have to memorize many inconsistent syntax and the rules of macro. Coding in SAS you spend 90% of the time debugging. The major advantage of SAS, in my opinion, is speed. That's the only advantage I can think of about SAS.
I agree with you. R is easier to learn compare to SAS.
Thanks Kunal for sharing your invaluable knowledge regarding this comparison. I live in Ontario, Canada. Can you please guide me about job market regarding SAS vs. SQL in Canada these days and probably one year from now in business analytics/data analytics? I have fairly good knowledge of SAS and SQL. I shall graduate after one year. I am doing my doctorate in Applied Mathematics from the University of Waterloo. In which directions should I proceed for certifications in SAS/SQL to be data analytics/business analytics? Thanks in advance, Best Regards, Puneet
thanks for information
i want to know a book on SAS for beginners
Nice articles. Great information. I have more that 20 years of industry experience (11 years in IT software alone in tech writing and testing) and want to move into Analytics. Given this scenrio, would you think "R" is good choice?
Good work !!! concise , crisp and to the point.
I completed my MBA (Finance) in 1998, for the past 16 years i am managing my own small scale plastic recycling industry. Now I would like to shift to Business Analytics. Is it worth to pursue PGCBA course from NIIT? Will my experience be counted? How about the job opportunities for me? Please suggest. Thanks.
Venkatesh, The course from NIIT has not received good reviews and would be waste of time. A better and far more effective thing is to look for online courses offered by Coursera (if you think, you can learn on your own) or Jigsaw Academy (if you need trainer assistance). Regards, Kunal
Hi, I am curious what about performance? Would be great to add this parameter. As far as I know R is very slow. Matlab is fast and Python somewhere in the middle (depend on used libraries). I do not know what about performance of SAS/SPSS. Someone know?
Hi! Try the R packages data.table or dplyr and compare it with SAS or Pyhton? R is very Low?
Hi Kunal' Recently i have completed my SAS course Base and Advance. Can you please suggest whether i can attempt the worldwide certification exam or i should get some experience before attempting. And please suggest some supporting courses which will help me in future with SAS. I can go for R or Ms sql Courses Like.
Hi sir, Recently i have complete sas and looking for job. i want to get suggestion from you that can i search job or i should some other course like R or phython;. And one thing more i can go for EXAM OR i should have to get experience. please suggest me
All are great languages. Having used Python, R, S, Stata, LIMDEP, and virtually everything out there over the last 25 years, when it comes to crunching out large scale production on complex data projects - and going home at 5 PM - SAS is by *far* the better choice. It's expensive, but so much of the dirty work is handled for you. It's not just a language but a complete processing system. There are stats procedures that are no naitive to SAS (very few) but you can program any estimator using the matrix language (IML) and seriously. If cost is a problem, check out World Programming System - WPS. They are a SAS clone language with basic SAS elements covered for a fraction of the cost. Cross platform to boot. Everytime I have to use these other systems I essentially cringe because I know it's going to take longer than I really want to spend. Life is short, the information value of a number is fleeting, SAS.
Which version of SAS are you using?? I am certain its not latest. SAS has best in class Graphic capabilities..! way way ahead than R. I use both I love SAS.
Zubair, I differ with your opinion here. SAS has pre-defined graphs, which are not flexible to changes. Try creating a heat map based on conditional formating in SAS and you will probably see what I mean. Regards, Kunal
I have learned about R but it seems that most of the companies are interested in using SAS. Also that i had Read some where written by the medical statistician that if they make some mistakes, or we havecome across some BUGS using SAS it will recompensate it but R does not have any feature like that....!!!
hi kunal I have a very little knowledge of c programming but i can manage with the basics. I have an interest in learning SAS. Since i have a poor background of c programming will it be hard for me to learn SAS?
Sana, Poor background in C has no bearing on the ability to learn SAS. Functional SAS is easier to learn (compared to C) and also has an option of using Graphical interface for a lot of operations. Hope this helps. Regards, Kunal
hi sir, i have 2 and half years old baby... so i didnt worked last 3 years. before 3 yrs i was worked in quantum, MR dimensions on market research. i learned SAS on that time. now i am confused which one i can choose. can anyone help me which tool i can choose?
Hi Kunal, The post and discussion here is very informative. Thanks for this. I would like to take guidance on what path to choose in Analytics. I have 3+ years of experience in MIS and data Analyst filed. Basically I am working in Excel, SQL and MYSQL. Which course I have to do, so It will give better Job Opportunity and How much cost I have to do for each and every course and also please tell me name of professional institute in noida providing SAS. Should I start with SAS or Python or R? Moreover, since I am more interested in consulting what courses should I go for? In SAS which version I have to used? Regards Sujit Sec-12, Noida.
Thank you for your post. It is helpful to have such a well-informed and balanced discussion of such a timely topic in an open forum. I appreciate the time and effort you put in to execute your careful review, as well as the time and effort you have put into moderating the discussion.
Hi Kunal, Nice article Great information Presently i am a Hadoop Administrator from past 1.5 Years, My Background is EMC SAN DMX/VMAX & Redhat Linux. I want to go into Data Analysis/ Data Sciences 1. How to become a Data Scientist? 2. Please guide me from where should I start (Technologies/Projects). 3. Is it mandatory to have a background into languages like R/Python/SAS/any other ( I feel comfortable with python), which one is most required as per Industry Demands? Look forward to hear from you, Shekhar Nagle +91 97666 97897 [email protected] Pune
Hi Kunal, very good scorecard, it should hold validity till mid 2015 then we should see lot of shift, with decision engineer's coming over data scientist and product as an outcome being more realtime than tool for analysis data i see a healty mix of open source and SAS/SPSS like tools, toards end of 2015 python will take the lead over R and SAS in terms of available people who know it, since web designing used it a lot and it is an engineering language. I have worked and tested about 50+ analytical softwares and worked for top 8 domains of service industry both as consultant as well as decision engineer. My vote clearly is for the software that would handle unstructured data well over time like, text, speech and finally video. or provide the best way of analysing and using it in prescriptive decision grids of PMML in real time. The cost of software should go down and the ROI should go up for Analytics by 4X or more.. Good read.. very interesting!!!
Hi Kunal, very good scorecard, it should hold validity till mid 2015 then we should see lot of shift, with decision engineer's coming over data scientist and product as an outcome being more realtime than tool for analysis data i see a healthy mix of open source and SAS/SPSS like tools, toards end of 2015 python will take the lead over R and SAS in terms of available people who know it, since web designing used it a lot and it is an engineering language. I have worked and tested about 50+ analytical softwares and worked for top 8 domains of service industry both as consultant as well as decision engineer. My vote clearly is for the software that would handle unstructured data well over time like, text, speech and finally video. or provide the best way of analysing and using it in prescriptive decision grids of PMML in real time. The cost of software should go down and the ROI should go up for Analytics by 4X or more.. Good read.. very interesting!!!
I think we are missing an important point here. While R and Python can be great for start ups and small companies that don't have a budget to pay for expensive solutions, I really can not imagine multi billion corporation using freeware for their risky business. Can you imagine a big credit card company running their Risk Management operation using R? I can't.
Hi kunal, the article is really good actually my confusio i am working with a Ecommerce company and its been 1 year now. I was from biotechnology background and shifted my field towards IT. My work here is to deal with quality and data base. So should i prefer sas for further growth or any other certifications. With growing market what you suggest in between the IT quality and business analyst. kindly guide thanku & regards Prakarsh mishra
Hi All, I have 10yr of IT experience including 6 yrs. in information access and enterprise search. I want to change or can say broaden my job area in Big data search and analytics. Just want to know from you guys, from where i can start to prepare my career in analytics. Thanks in advance. Mithlesh
Mithilesh, You can look at some of the courses from Bigdatauniversity to start with. Or undergo Cloudera certification. Both of them can give you good start into the domain. Regards, Kunal
Mithlesh, Would recommend learning Big data Tools (Hadoop/Pig /Hive/MapReduce) to expand into Big data Analytics. You can look at Wiley certified Big Data courses or the course on Edureka. Regards, Kunal
Hi Kunal, First of all i would like to thank you for the valuable information given above.I am working in a ecommerce industry from the past 2 years(basically into operations) and i would like to learn analytics.I would need your help to sought out few doubts i have. 1) Is it a good career option to learn analytics? 2) From your feedback provided above i believe u insist learning SAS
Hi Kunal, First of all i would like to thank you for the valuable information given above.I am working in a ecommerce industry from the past 2 years(basically into operations) and i would like to learn analytics.I would need your help to sought out few doubts i have. 1) Is it a good career option to learn analytics? 2) From your feedback provided above i believe u insist learning SAS first and then go for R language.So should i go for a certified business analytics professional programme and then go for a certified R programmer course? 3)I have a 0 knowledge in analytics and a basic knowledge in SQL and C.Basically i need a job change and want to grow my career as a BA. What would be helpful.Kindy suggest. Look forward to hear from you, Abhijit
Hi Abhijit, 1. Yes, It is expected to be one of the high growth sectors in next 5-10 years. A word of caution is that employers are currently looking for people with some experience. 2. At this stage, I would suggest you to focus on a single language – SAS. You can do a certification course of Business Analyst with SAS then apply those learning to a few problems. 3. Apply your learnings to a few datasets and participate in a few Kaggle competitions. Gain hands-on experience and approach to the job openings from our job portal. If you do well in Kaggle competitions, you can get good job opportunities. Regards, Kunal
Great comparison done! Thanks for sharing also! The Data Science function is not fully formal yet in many large F100 / F1000 enterprises. There are few roles starting to appear on both business & IT domains of Data Science. Hence expecting to find a comprehensive enterprise class Data Science stack will be really hard. There's lot of hype and glamour surrounding Big Data, where both commercial and enterprise Hadoop stacks have been deployed and playing a successful role. Even the Big 3 (IBM, Microsoft & Oracle) are working on their flavors of Hadoop either organically or through partnerships. These Hadoop stacks enable conveniences for languages, connectors, APIs, storage management, DR, High Availability, High performance, real-time vs batch processing SLAs etc. But there's little or no emphasize on analytics / statistical modules bundled with them. Few exceptions are Mahout, MapR, but mostly a standard based open source extension. As an Enterprise Architect, I prefer R (R Studio) for EDA with a more collaborative context on Git / SVN primarily for Data Scientists / Statisticians / Data Modellers. Even though I'm yet to do this, we are actively looking into Cloud based Hadoop stacks which has native support to run R large scale on Big Data sets. From an analytics consumption perspective, we are exploring Tableau (can really augment R graphs well) and our current landscape (MicroStrategy & Microsoft BI - Fancy Excel).
Hi Kunal, I have done B.tech in Electrical and Electronics and have a experience of more than 1 year in Market Research field. Now I want to go in Business Analytics and i'm interested in doing a course in Analytics. So please suggest me a good training institute in Delhi NCR. Regards, Ashish Gautam
Hi Ashish, If you want to go for certification in Business Analyst, you can undergo the certification from SAS Institute. SAS Institute provides trainings for these certifications in Delhi, Bombay and Banglore. However, if cost is a constraint, you can do freely available online trainings. You can also look at our SAS learning path here. Regards, Kunal
Hi kunal, I am working as data analyst having 2 years experience but with no knowledge of SQL. My educational background is B-tech in electronics and communication. I have opted for R and hadoop language programme. So please suggest how it will help me in boosting my career seeing current industry trends. Also suggest if I should have done SAS instead. Thanks in advance. Please reply.
Hi Ayush, As you have mentioned in the question that you have opted for R and hadoop. If you have started learning R and Hadoop, then I would suggest you to proceed. As Foundation course in SAS is the basic course. If you haven't enrolled yet for course in R and Hadoop. I would recommend you to move forward and learn foundation course in SAS rather than R. Regards, Kunal
Dear Kunal, Sorry I am visiting your blog only now - regret how I missed it all these days! Very informative and current. By the way, I am working in IT of a large PSU for over 20 years. I am working with JSP with Oracle backend. I have worked on Data mining techniques. I am getting interested in Analytics, in a bid to analyse some of our business data (no official investment though). I am doing R with Coursera as I heard that SAS is a priced package, although said to be more comprehensive and feature-rich than R. However, I wish to know the following: If SAS is easier to learn and use, then can I have the link to download a free version of SAS (eg., Base SAS)? Pl. provide the details.
Dear Kunal, Also, I'd like to know how far I can use SQL with R to handle data as we are already using SQL with Oracle?
hi at present iam doing msc-statistics final year,and also iam learning sas, so can you tell how be the future and opportunities for me
Nice one Kunal!
I would still go for R any day any time. It has high functional capabilities.
Hello Kunal! I am a B.Tech in Chemicals Engg and a MBA with marketing.My entire career for 11 yrs till date in in Sales and marketing for specialty chemicals and bulk chemicals used for industries like Paints , Construction Chemicals, Adhesives etc. For the last 2 years I am working on forecasting of sales , conjoint analysis for new product development, penetration rate of existing products for my organisations apart from sales. Mean while I ve bagged cirtification on R Programming language and Business Analytics. Frankly speaking I am getting bored with this Sales job and want to concentrate fully on Market Research using Business Analytics. Can you provide me some light as guidance???
Hi, I am currently doing mba financial services from a tier 2 B school in Mumbai;posses a BE (IT) background with experience in pl/sql programming in a MNC.I am interested in Credit Risk analytics like modeller or model appreciation etc.Currently I am learning R;but I doubt whether R is used in this segment or not.Would it be better to learn SAS.
Hello Aniruddha, R provide good support for Credit Risk Analytics. Using Logistic Regression or CART in R you can differentiate between “low risk” and “high risk” debtors prior to granting credit and classify them as default and non-default and identify the key determinants of default behavior. On the similar line you can also develop a predictive model in R order to identify the propensity of the accounts to pay or not pay their bounce EMI within duration of one month. Only the problem is that you need to write your own R code/R Program. Regards, Amol
Dear Kunal, I am B.Com Graduate having around 7 years of experience in IT Recruitments.For the last 3 years I have been at home doing Freelance Recruitments.Now the fact is I am not getting enough leads from my associates and tell u honestly my life is at stake.My currents earnings are almost NIL. Now my plan is get trained in Data Analytics with R and want to switch my career as DATA ANLYST. If I take up this course can I get a job at this point. Pls help...thanks.. gvrp
Gvrp, Which course are you referring to? Regards, Kunal
Hi Kunal, I was thinking of taking R programming, but just came to know about SAS and not sure at this point which way to go? I live in california and would like to get in medical industry and would stay there for next 10 years. I am 55 and not too ambitious at this age. guide me with your expertise which way to go . R is difficult and not sure about programming. I do have background in computers and worked in companies as QA engineer. thanks
I would like to know also from which site I can do the course and how to proceed.
Hi Kunal .. Such a nice info you have here and others too. Need your help to understand following few points. 1) I have been working in Supply Chain industry for past 2 years 2) My profile is mostly into B2B and Channel sales 3) As I have a technical background and have interest in Analytics I want to change my domain 4) Very recently I came across SAS Certification, so I'm planning to do it 5) Considering my exp. sales what do you suggest SAS or R , I also want to add that I dnt liberty of leaving present JOB and do certification and serach for job.. I want something which I can pursue while doing my job 6) If SAS then which module should I opt for ? I could see many modules in SAS .. plz advice.
Hello Kunal, Such an amazing read! I need some suggestion. I have worked as a Java Developer after my Masters degree for a year. Then unfortunately I have had a gap of 4 years. I am looking to enter into the job market now. But with my little experience, what should be the best path to enter into the job market? I have been looking into Certification courses in Business Analytics, Big Data, also many courses in R Programming, SAS, MicroStrategy, Hadoop. A few friends have suggested MicroStrategy. What is your take on it considering I am from Delhi? Thank you in advance. Julie.
Kunal sir, I am a b.tech grad (fresher) and want to start a career in analytics. Which certification shall i take and from where...as well as what are the entry level chance here.. Hope for a helpful reply. Thank You
I have used both R and SAS so far . But found R to be much simpler to adopt. Being an open-source tool it has end-less possibilities . Probably my views might change after I learn Python ..
R is not a low level programming language, it's high level - I'd even say it's higher than Python or SAS because you don't have to define data types. This is part of the reason it is slow. Also saying SAS is easier to learn because of the SAS SQL Proc seems odd - SQL is easier to learn, not SAS and R has the SQLDF package so you can write SQL with R and pandasql with Python in the same way
I am too young in this field to contribute to this discussion but i must admit that every word on this discussion page is worth reading for some one like me who is planning to start a career in analytics. Thank you Kunal and all the participants
Hi Kunal, Thanks for this post! I got what I was looking for. I have a total of 5.5 yrs of experience in Mainframe Operation. There is no programming involved in my job profile. I have decided to move on to the Analytic's side. Can you please help me out to choose a better one out of it from the job perspective and ease of learning, considering my past experience. Will the market accept a person with no experience and only certification on SAS? Thanks and regards, Nikhil
SAS, Python and R are useless in Derivative Modeling. C++ is King in Modeling Derivatives.
Matlab is still awesome! I love it!
Nice Post, Now these days SAS is more popular than other one.
Hello everyone and special thanks to Kunal for above discussion, it is really help full for beginners like me more over most of my doubts are clear but still I need some advise I have 9 years of experience in technical area only using one of the legacy systems now I want to switch to another technology for that I had decided R+SAS but where ever I going for the enquiry every institute is giving different picture I want to be at technical side only here is my confusion do I need to learn sas data science along with R+SAS or it is enough if I will learn R+SAS technology,please do advise this will be great help for me.
Well Said!! Thank You kunal Sir for the comparision. I have found the same comparison on a lot other websites. SAS is leader. But is there any way to learn SAS on beginners level and somewhat moderate level for free, as the training by SAS institute at pretty steep
You get what you pay for$$$ SAS wins!!! in most areas: Big Data #1 - Pages and doesn't run in memory so it doesn't run out of memory and crash Stats #1 - Verified by FDA its a standard, R contains a lot of junk written by "anyone" and not verified, over 45 regression packages ! which is correct to use? Visual Analytics #1 - R and Python does not exist Parallel CPU - SAS , R is just starting on H2O.... If you want a little school or small business analytics then go for free stuff like R, Python, Octave, the list is long.
Also, SAS is free to learn, download it and see: http://www.sas.com/en_us/learn/analytics-u.html
Hi Kunal Did you notice that this article is a copy of your article? https://www.linkedin.com/pulse/sas-vs-r-python-carlos-r-bueno Cheers Hitesh
Thanks Hitesh for highlighting this. I have already informed the Linkedin team about it. Hopefully it would be actioned quickly. Regards, Kunal
Hi, I'm pursuing a management (analytics specialised) course. I'm interested in financial sector. Which Tool should I go for? This is my first year of the two year course. I want to get a good knowledge of the analytics tool, before going for my second year. How long will this take to excel in any tool. If you can provide the links for free tutorial for R, that'd be really helpful. P.S. I've sent you request on ur linkedin. Pls accept.
Hi Kunal, I am working as SAP ABAP developer with 2 years of experience. I want to move to Analytics side. So, is there anything I can do in analytics so that my SAP experience does not get waste ?
Hi All, I have 10+ Years of experience in Oracle Products. (SQL/PLSQL,Forms,Reports) with PMP certification. Now, to increase my salary i am looking out for options. Found Data analyst is good. Now, i am planning to switch (carrying my previous experience) to Data analytics – Really confused on where and how to start ? Should i go with SAS or R ? Also if anyone can suggest which will be more suitable for me ? Learning Path ?
hi kunal sir My self saurabh kumar I completed my B.E in stream of information science in the year 2015 with 50% I am interested in doing course on SAS or R programming.. kindly suggest me which course is better for me as fresher, and in which field more job is avaliable kindly suggest me the good institute for SAS OR R programming, IN MUMBAI OR BANGALORE THANK U [email protected]\
Dear sir My self saurabh kumar I completed my B.E in stream of information science in the year 2015 with 50% I am interested in doing course on SAS or R programming.. kindly suggest me which course is better for me as fresher, and in which field more job is avaliable kindly suggest me the good institute for SAS OR R programming, IN MUMBAI OR BANGALORE THANK U [email protected]
Dear kunal, I am Ece graduate with no work experience ,I want to enter into data analytics field, I don't have any coding experience ,I know C, What languages should I learn for a beginner like me,and to get job in the market
Start with R. Even though Python would be easier and language for future, currently there aren't many jobs on it in India Regards, Kunal
Dear Kunal, First of all i would like to thank you for above article and would need some suggestion here. Currently am working as QA engineer with 2 years of total experience. There is no programming involved in my job profile( 2 years back I learnt Java, but dont have work experience on coding ). Now I have decided to move on to the Analytics side for better opportunities. should i start with R or SAS ? Kindly advise based on my current experience.
Hi Kunal...recently I came across this "Tibco Spotfire" analytics tool..how it is & what's the future of this tool. ? Regards Mayur P.
Hello Every One, Feeling to good to be a part of this discussion. I would like to add "Odoo". Odoo is ERP created using Python and is giving a strong competition to other ERP like SAP, R, MATLAP etc. Regards, Kuldeep Singh
Just a high level overview from my experience. I started out using SAS from university to my first job. Then got into R and experienced it's power and started not too long ago with python. Doing my research last year on this topic it was clear that SAS is dying... no longer in universities, too expensive and also not a wise choice for startups. SAS seems to cater mainly for credit risk analysts.... period. R and python however goes way beyond and into Data Science and many other opportunities. Not to sound harsh but in my view SAS shouldn't even be in competition with R and python. My primary language is R at this moment but I'm having more fun with python and I guess it will become my favourite as python feels more powerful iro it's diversity and applications. My advice to all Next Generation analysts is R or Python! Learn them both. After learning R first python become much easier to understand.
R and Python are opensource frameworks, Eclipse provides the ability to download plugins for both.once development platform for both , hence I don't have to haggle around with IDEs for development , I guess that's the beauty of using R and pythan
Great post. Thanks for sharing.
Hello sir I did my pg in statistics and then did mba in finance and marketing. I m too much interested in number or quantitative subject .right now I confused which course I suppose to do sas. ...r....data or business analytical anything else pls suggest me and guide me.
hi sir, I did BE in INORMATION TECHONOLOGY . I am a fresher looking forward to make careeer in analytics. I was not much interested in IT jobs therefore never tried to learn something new, but now as i heard a lot about analytics i am keen to know about it in depth and want to get started. sir i request you to please give me an advice about what language to learn,which one will be suitable to get job. I am moving to pune so please suggest me institutes from where i can learn the languages.
Hi Kunal, I don't have coding background. Between SAS, R, Python which one is easy to learn. Please suggest. Thanks
Dear Kunal, I would need some suggestion here. Currently am working as process associate In accounts payable with 3.8 years of total experience. There is no programming involved in my job profile( I don’t have work experience on any coding). Now I have decided to move on to the Analytics side for better opportunities. I have done my MBA (finance) in the year of 2013. From 2013 I am in to the account payable. should I start with R ? Kindly advise based on my current experience. should I step into analytics side, is it OK ??? Thanks & Regards, Shanigaram Thirupathi
If you have a strong firewall, and if you cannot load packages Python and R become quite inefficients...
Considering the overall score, it seems that python is a better option for data science community. Nicely explained
Thank you for the above nice article In my opinion, within a couple of years MATLAB will overtake SAS, because it is cheaper than SAS and machine learning, deep learning support are available whereas SAS doesn't have this kind of support.
How about licensing system Most of the "R" CRAN packages are coming with GPL Licence (GNU License) with Copyleft option where as Python packages are coming with Apache, MIT or BSD License
Have read the 2014 version of this and can now see the massive move in market. As a Analytics beginner with SAS, this article was very helpful.
Hi, The article is very useful, i am a fresher and working as data scientist since 3 and half months, i know all 3 tools but i am very very good at python and have good hands on python but i am average in R. do you think i need to learn R or python is ok for me to carry on my career as data scientist ? Need your suggestion Thanking you in advance.
Prophesies of the death of SAS are like Mark Twain's famous quip "the reports of my death are greatly exagerated." Here we are some 3 years after the projections made here of the death and/or decline of SAS and SAS is still the dominant analytics platform. I started learning SAS some 34 years ago (my first language in school was Fortran using punch cards). Over time I have learned and usesd professionally some 33 languages. And I'm currently working on python and R. I work in a forecasting department of some 30+ forecasters (largest forecasting group I've ever worked in). I know of no corporation that is reducing their use of SAS and certainly none that are switching from SAS to some other language. There may be some out there but I don't know of any. When we are hiring, analytic skills (econometrics, math, stat, etc) are dominant and if the person already knows SAS then that is preferred. Other skills can make the person even more valuable. Java, HTML, CSS, python, etc. (We don't even have R installed on our server.) SAS is not dead and will not be dieing anytime soon. Why? Because it provides comprehensive data handling, well tested and dependable basic to advanced analytics and, in the hands of an expert, a rapid development cycle. It doesn't yet make machine learning as easy as python does (which is why I'm learning python) and it has some other minor flaws but what language doesn't? That's my 2 cents worth based on 34+ years experience.
Hi Kunal, I am working in an IT company as Business Analyst. I would like to know which course/certification would be good for me to keep my job secure in IT industry. Could you please guide?
As per the recent KDnuggets poll on R vs Python. It has been shown that Python has overtaken R in data analytics. look at here http://www.kdnuggets.com/2017/09/python-vs-r-data-science-machine-learning.html http://www.kdnuggets.com/images/google-trends-python-data-science-r-2012-2017.jpg
I actually think R's steep learning curve is overblown. The fact that there is one widely accepted IDE (RStudio) is immensely helpful. Installing packages through RStudio is a breeze. Python may lead in deep learning, but Keras and TF are now available in R. All other ML tasks are easily performed in R as well. Anyway, end of story, you should work on becoming proficient in R and Python! They are both amazing languages for data science related tasks, and only continue to improve and grow.
Hello Kunal, Thank you for the article. It is very nice, clears a lot of doubts. I am a 3rd year student pursuing btech. In today's date if you compare these languages what will be your opinion? And please share your opinion regarding the 2019 industry trends, what do you think which language will provide good jobs in India and outside also?
Well explained! just thought to share one thing with this community. We can access free SAS online compiler on web without even need to install it in your machine. I am practicing SAS with that only. As I have started just now to get my hands dirt with it, I am not able to say its pros and cons right now
Interesting - do you have a link to their free online compiler?
I think ease of learning should not be a criteria for comparison. If a particular tool is useful but difficult to learn, a professional should go ahead and learn it rather than choose an easier option.
I am a very experienced SAS programmer needing an open source option for some free lance work. What's my best option for an easy learning code?
Thank you Sajan, this is good overview.
Hi! I know the SPSS & STATA, Could anyone suggest me that which programme would be the best to learn easily and getting ongoing opportunities? Thanks
I'm dealing with SAS right now and I think the "ease of learning" thing is way off. SAS is actually horrendous, requiring the mixing of their macro language with "regular" code to do anything remotely interesting. I won't say that R is easy by any stretch, but to actually do anything of even modest complexity, I'd rank R as easier than SAS. Python is the easiest of them all... I would have SAS = 1.5, R = 2.5, Python = 3.5 for that. None are a breeze, but clearly Python > R > SAS. Also for "Deep Learning Support" you're spot on, but I would even generalize that to "Cutting Edge Development." For instance, XGB was developed in Python and only then ported to R... GPUs are supported in Python with R only beginning to take baby steps... and of course Deep Learning which you mentioned. Cheers.
PS, Administering SAS is a nightmare, too. Granted, the analyst needn't worry about that, but just saying... their licensing structure is a well known quagmire. Integration with source control tools is much more robust in either R or Python common IDEs (RStudio, Anaconda, pyCharm, Visual Studio, etc...). Maintaining a server environment with RStudioServer or Anaconda Enterprise or Microsoft ML Server... all much more straight-forward than SAS. But anyway, not to totally bag on SAS... I agree their tech support is good both officially and at their community pages.
Great Post and Very Informative discussions Question on the point "Published about 3.5 years ago and updated a month ago" Is there a graphical representation of how these scores have changed over these 3.5 years or year on year? Are these the same attributes applicable today? "Integration with other tools/softwares": Isn't this another attribute to compare all options? With the integration/support of R on platforms like SAP HANA and Tableau, does R have advantage over Python?