dterpsss
SOHH OG Class 2004
Can I teach myself easily on udemy or do I need a real class? I want to get to an intermediate level for each, I want to start with the easiest one to digest, thoughts?
Anyone gone through the process of learning tableau or R?
- Thread starter dterpsss
- Start date
Dabbled in R a little bit. There is a package in R called Swirl (you can read about it on swirlstats.com) that teaches you how to use R inside of R. Very helpful.
With that said, I kinda fell back on R because the main use for R (in my company at least) required a strong understandings of Statistics. It's been a while but IIRC, the main purposes of R was to rigorously analyze data for meaning while Tableau is a way to easily present data once it has been analyzed via dashboards et al.
I think Tableau is the most accessible and utilized in the everyday world, but R is the most powerful (when you know what you are doing and why you are doing it).
I would start with Tableau and play around with R on the side via the Swirl package to see if you like it, but know that you will need to expand on your statistical skills in order to take full advantage of R.
I fooled around R but its closer to programming. Tableau is like excel on roids.
I know R. The more statistics background you have, the more you're going to get out of it, as mentioned before. I haven't done anything with Tableau, but everything I've seen makes it seem like it's a colorful pivot table at the end of the day.
I honestly use more python/pandas/scikit-learn nowadays, but R is still good for more straightforward "mathy" problems.
To actually answer your question, it's really dependent on how much you leverage your existing stats knowledge. It's not a particularly hard language to learn, but to make full use of it, it's assumed you know why you need it instead of a more general language.
I honestly use more python/pandas/scikit-learn nowadays, but R is still good for more straightforward "mathy" problems.
To actually answer your question, it's really dependent on how much you leverage your existing stats knowledge. It's not a particularly hard language to learn, but to make full use of it, it's assumed you know why you need it instead of a more general language.
Last edited:
dterpsss
SOHH OG Class 2004
i know basic stat ...hypothesis testing, reg analysis, probability, some model selection crap etc
dterpsss
SOHH OG Class 2004
Im in Bschool trying to get an edge
dterpsss
SOHH OG Class 2004
delete
ByAnyMeans
Rookie
I know both software's really well. Ive built applications in R and done a lot of data wrangling and some statistics in R. I use SAS for my stats work primarily. I want to do more Stats and machine learning in R though.
But to me this is easy. R is free whereas tableau is not. Learning R will be more programming and tableau is simply point and click.
I've solved a lot of business problems for people in R.
If you want to easily present data tableau is the way. But if you know how to program well. I can create what you do in tableau in R and make a web app opposed to having a tableau license. The real question is what the hell do you want to do with the knowledge?
If you see yourself leveraging it daily probably R. Because I don't think tableau is powerful enough and to me r gives more freedom.
But to me this is easy. R is free whereas tableau is not. Learning R will be more programming and tableau is simply point and click.
I've solved a lot of business problems for people in R.
If you want to easily present data tableau is the way. But if you know how to program well. I can create what you do in tableau in R and make a web app opposed to having a tableau license. The real question is what the hell do you want to do with the knowledge?
If you see yourself leveraging it daily probably R. Because I don't think tableau is powerful enough and to me r gives more freedom.
dterpsss
SOHH OG Class 2004
how so? modeling?, aside from the extra skill on the resume solving business problems is really my motivation
ByAnyMeans
Rookie
Taking in a lot of data and visualizing it with dashboards, assessing patterns, haven't done much modeling with it but some people at my company have. Descriptive analysis for the most part. Data wrangling too. All the modeling and stats are normally done in SAS first on the projects I worked on. And then everything was dumped into R using shiny to visualize it.
If you think you'll have a lot of data then I think it will be advantageous. What type of business problems are you getting into?
This is just wrong. If I need to do nonlinear least squares, linear mixed effects, or even rolling my own regression with a custom error function, R is more robust and obviously the tool for the job.
Even for simple things like ANOVA, R builtin is better than Python's statsmodels.
dterpsss
SOHH OG Class 2004
I'm in business school, going to work at a silicon valley company for my bschool internship, mainly in logistics/Demand analysis