Reply to: SAS - help!

[Bagpuss]

Nah. SAS is stats with "wizards". I've seen it in action. It's not all that it's cracked up to be to be honest.

Sounds like you are on about SPSS not SAS

[sasguru]

Nah. SAS is stats with "wizards". I've seen it in action. It's not all that it's cracked up to be to be honest. Matlab would be better for the job in most cases IMO, but it's regarded as "academic" rather than "corporate".

Wizards? Wizards!
Clearly you've been using EG or the Enterprise Miner interface. That's for amateurs, and they do do more harm than good if you don't know the assumptions and principles underneath.

[bored]

Nah. SAS is stats with "wizards". I've seen it in action. It's not all that it's cracked up to be to be honest. Matlab would be better for the job in most cases IMO, but it's regarded as "academic" rather than "corporate".

It depends. Matlab aint that great when you have to process tens of billions of transactions...

[Cowboy Bob]

Ooh, I hope you're not frightened of maths and that stats always brighten up your day....

Nah. SAS is stats with "wizards". I've seen it in action. It's not all that it's cracked up to be to be honest. Matlab would be better for the job in most cases IMO, but it's regarded as "academic" rather than "corporate".

[DimPrawn]

.NET does advanced maths.

+ - / * % that kind of thing.

HTH

[Flubster]

TBH I thought it was a database a la Oracle but from the above posts I guess that's not quite right...

SAS has it's own proprietry database, but can store/read data from practically any other data source. No reason why they couldn't read data from DB2 then store in Oracle, but then that would be like giving you a speedboat and saying you can't turn on the engine, but here's some oars.

SAS, like BO (spit) are trying to deliver end-to-end solutions from data warehousing through to MI delivery...

HTH

[sasguru]

I realise that, but that's its strength - there are other languages out there that can do bog standard stuff better.

Wrong - nothing does bog-standard data analysis better.

[cojak]

TBH I thought it was a database a la Oracle but from the above posts I guess that's not quite right...

Oh Dear(tm)

[gadgetman]

Hmm, what have I started here?

As I mentioned, I am in Test. There are loads of SAS developers beavering away here.

I am going to be testing some reports and the configuration of some off-the-shelf SAS tools (i.e. I'm not testing the tools), which is why I thought in may be helpful to do some offline research.

TBH I thought it was a database a la Oracle but from the above posts I guess that's not quite right...

[cojak]

I suppose it depends what the OP will be using SAS for. It's not just a stats package, don't you know.

I realise that, but that's its strength - there are other languages out there that can do bog standard stuff better.

[Old Greg]

a competent and smart programmer

will stand out like a sore thumb here. Let's assume the worst.

[Flubster]

Aye if he's doing data step programming, a competent and smart programmer with other language experience can pick that up in a week.

Or just use PROC SQL and take out any non-SAS specific code...

Or for the real SAS newbies, use EG.

[sasguru]

Aye if he's doing data step programming, a competent and smart programmer with other language experience can pick that up in a week.

[Flubster]

I know you do SG, I was talking to the OP...

I suppose it depends what the OP will be using SAS for. It's not just a stats package, don't you know.

[cojak]

I know you do SG, I was talking to the OP...