Warning: Creating default object from empty value in /home3/science4/public_html/rconvert.com/Convert_SAS_To_R_Statistical_Data/wp-content/themes/optimize/functions/admin-hooks.php on line 160

Should I Switch to R?

The answer to the question, “Should I Switch to R,” will likely be different for various users & businesses.  While many companies may feel comfortable with their current statistical and data computing software, there is strong motivation for others to change.  To determine if one should switch, listing out pros and cons, then calculating a Return on Investment (ROI) and its associated timeframe can be helpful.  Below are some thoughts about what may be considered:

Pros

  1. No more licensing costs
  2. Large user community
  3. Powerful database integration
  4. Ease of use
  5. Integration with the web and other technologies

Cons

  1. Size/complexity of current environment.  Large / complex = more expensive
  2. Employee skills – will new staff be needed?
  3. Conversion know-how / time.  Do we have the skills to switch, and how long will it take?

Our experts can guide companies through each of the concerns that may impact a decision of whether to switch.  Feel free to contact us if you would like some guidance.

 

guidance
0 Comments

10 Reasons to Switch to R (from SAS, SPSS, and other data software)

In our experience converting businesses from SAS to R (and SPSS to R), we have compiled a list of common reasons why customers make the switch to R.  With R gaining such attention in mainstream media and international business,  more companies are questioning whether converting to R would make sense for them.

Here are some of the reasons why our customers have converted to R:

  1. R integrates with a large number of database systems.
  2. R supports parallel processing.
  3. R can support huge data volumes – gigabytes of data and 100′s of millions of records.
  4. R implements new methods faster than any other data technology.
  5. R is cross-platform.  Windows, Unix/Linux, Mac, etc..
  6. R has a huge and active community of smart people.
  7. R has a powerful and consistent base language – if you dream it, there is likely a way to do it in R
  8. R is well-documented (look at Amazon.com!)
  9. R integrates well with other languages and web platforms.
  10. Best of all, R is free. Never worry about losing your work for not paying a bill.

For more information on converting to R, visit Rconvert.com.

 

0 Comments

SAS and R – A love/hate relationship

Once upon a time, I was the manager of the SAS department for a leading SAS partner in New England. As a SAS aficionado who’s work with the language dated back to my days in college, I truly valued the power that SAS brought to the table, whether by providing data management, statistics, predictive modeling, or business intelligence capabilities. My expertise was centered around using SAS technology to predict customer behavior for a variety of industries – insurance, marketing, and healthcare in particular. SAS is a fantastic tool for data mining, predicting, and forecasting in these areas.

While the technology and its application were solid, I began realizing that lots of small and mid-size companies were having difficulty footing the steep SAS licensing bills. One of the major disadvantages of using SAS was the annual licensing fees. Anyone in the field knows that these fees are not trivial. While large corporations like a Bank of America might be able to pay these costs without issue, the small to mid-size businesses were getting left out. Entrepreneurship, in particular, was being stifled by these overbearing fees. Even more scary, if a company begins developing in proprietary languages licensed annually, then hits into rough times that make it unable to pay for renewal, all the developed work could be useless since the company’s license to use the product would turn off. Talk about a major corporate risk! Would you build your dream house on annually-leased land?

While my career was still focused on SAS, something else was happening in the data & analytics field. R, a language that had once been dismissed by all but the advanced academic, was starting to make its way onto the mainstream business stage. The origins of R date back to the work of John Chambers at Bell Labs in 1975.  While at Bell Labs, John formed a statistical computing language that later became known as “S.”  Around 1993, R was written as an implementation of S by Ross Ihaka and Robert Gentleman at the University of Auckland, New Zealand.  R is open-source and freely available.

In the early 2000′s, I had done some experimentation in R.  Compared to other tools I was using, including SAS, SPSS, Matlab, and Stata, I was not too impressed with R.  I found that it frequently crashed and seemed inconsistent at times. This led me to pass on using the tool and not take a second look until nearly a decade later.

Frustrated with the challenges I saw with proprietary tools in recent years, I decided to take another look at the available technologies in the field.  What I found shocked me.  R, a tool that I had dismissed years prior, had come a very long way.  In fact, once I began using R, I could not put it down!  After about 4 months, I realized that I could do everything in R that I was doing with SAS.  Best of all – R was completely free!

What I really like about R is how quickly the R community has built the tool up.  I’d estimate that R grew more in the last 3 years than many similar proprietary tools grew in 30 years.  That is the power of open-source, that is, there are a lot of eyes on R.  Comparing R to other similar technologies is like comparing Wikipedia to an old-fashioned encyclopedia. In terms of capabilities, I have come to believe that R will become the de-facto standard in business statistical computing over the next decade.  From its database integration to its parallel processing abilities, R is a fairly complete package.

After realizing how far R had come, my eyes began to open about who else was using R.  Some of the top technology companies of our time- including Google and Facebook – are R users.  As fate would have it, around the time of this realization, I received a call from the representative of a large Wall Street financial firm.  They needed help converting a SAS environment over to R.  At first, I thought this was unusual, but I then realized that this is what other large companies are doing.  Hence, Rconvert.com was born – to help companies make this transition.

 

 

0 Comments