# file wanted

#### Hamid

##### New Member
Can anyone send the file cluster.sas to my account hamid196509@yahoo.com, or copy and past the contents of it to my inbox at Talk Stats please. Thanks.

#### Dason

I've never heard of this file. Is this for a class? I mean I have a file on my computer called cluster.sas but I guarantee it's not the file you want. In other words: give us more detail.

#### Dason

No. It is not the file you want. I guarantee it. Want to know why? Because I'm the only person that has ever seen that file. I wrote it.

You may think that cluster.sas is a well known file or something. But just answer my questions that I posted earlier.

Is this for a class? What do you expect is in this file? Give us more detail in the next post or else I will not respond anymore.

#### trinker

##### ggplot2orBust
Dason said:
In other words: give us more detail.
We have no idea why you want this information. For all we know this file is a file that contains all our data we've ever entered into SAS and giving you this would give you access to our confidential data. IRB would be argry for unethical distribution of data, we would lose our job and in 6 months we'd be homeless. It's too risky to just give out his information to someone who was requested to:

Dason said:
I've never heard of this file. Is this for a class? I mean I have a file on my computer called cluster.sas but I guarantee it's not the file you want. In other words: give us more detail.
..and responded:

Hamid said:
Yes this is the file I want
If you want your request to be fufilled it would be prudant to respond to the request of those who have what you want. Choice is yours but I'm betting I'm the last member response to this post until you provide the information about this file.

Last edited:

#### Dason

Since I'm such a nice guy I decided to upload a slightly different version of my cluster.sas file. Enjoy! To upload it I just renamed it to cluster.sas.txt but don't worry it's still the same awesome file!

View attachment 1505

#### Dason

What can I say? I'm just a nice guy...

Did you have a chance to check the file out? It's pretty awesome and does a lot. I bet it's exactly what they were looking for.

#### trinker

##### ggplot2orBust
Actually I'm not a SAS user. It scares me. I have to learn it with a regression analysis class this Fall. I hope it's not too difficult to learn (I really like my R but being proficient in something else would be an asset).

#### bryangoodrich

##### Probably A Mammal
If this is a regular SAS file that comes with the software, there is a simple reason to deny Hamid's request: it can be seen as a violation of the TOS. All the intellectual property of SAS belongs to SAS and is permitted to those with a valid license. But if Hamid had the software, then Hamid should also have the file, making the request moot. Therefore, it seems odd to just ask for some random file as if everyone has it, and that Hamid should be given access to it. Now, don't get me wrong, I'm all for file sharing, but I would never just hand it to anyone just because they said "please" and "thanks." I just love the "yeah, that's the file I want" without even knowing what someone else has! Apparently any file named "cluster.sas" will do! lol

#### Dason

I'm pretty sure the file I provided will solve any problem ever posed.

Plus I don't think there is a file called "cluster.sas" that is part of the SAS software. It's probably a file they need for an assignment or something and should have been provided by the professor. But we won't know until they tell us more.

Or maybe the file I provided is exactly what they need.
But I doubt it since it was just a file with a SAS style comment saying
Code:
/* This is the file cluster.sas - I hope you like it */

#### Hamid

##### New Member
Dason,
Where is your mom?. I want her in my bed. Make sure next time you do not joke with people, otherwise your mom will be insulted again.

#### Dason

Dason,
Where is your mom?. I want her in my bed. Make sure next time you do not joke with people, otherwise your mom will be insulted again.
And why can't I joke with you? You've provided no information to help us here and your request is laughable at best. I'm sorry. We'd love to help you but you're making it impossible. I provided the best file I could find (think it wasn't a good file? It's a runnable SAS file named cluster.sas - which is actually better than what you've told us because I don't even know if what you want is supposed to be runnable!). I know it wasn't what you wanted but you're being ridiculous here. If you actually give us some information to go on we'll provide you with help if we can.

Also - Insult anybody or their mother again and consider yourself banned. You're lucky that I didn't ban you outright for being ridiculous and jumping to insults. I feel if I did ban you I might regret it later for acting too harshly so I'm restraining myself for the time being.

If any of the other mods feel this is bannable then you have my support in kicking them from the site.

#### Hamid

##### New Member
Dason,
I am sorry for insulting you. It was insane. I feel so ashamed for it. I hope you will forgive me. If you do no want to help others (for some reasons justifiable to you), then at least do not joke with them. People may react differently and unexpectedly.

#### Dason

The thing is that we want to help. But we need you to tell us what you actually need. Have you not been reading? We've been asking you questions and you've failed to answer any of them. If you answer some of these questions we might be able to help. But with the information you've provided we can't do anything.

#### Hamid

##### New Member
Thank you Dason,

I am studying SAS for cluster analysis, and I want to see how it handles outliers. I need to look at the actual code to see how it tracks the outliers. Unfortunately I did not find any document that would clearly say how SAS handles outliers. They just say that SAS uses the TRIM option in PROC CLUSTER to find outliers; but I do not know the theory behind it. It would be as good if I receive a documentation that would explain the theory as well. I sincerely apologize for my insult above, and seek your forgiveness.

#### Dason

halfnormal gave good links. SAS isn't an open source program so you're not just going to be able to look at the code it uses. The help files for SAS explain what the procs do though so that's a good place to start (halfnormal linked to the SAS help).

#### trinker

##### ggplot2orBust
I see Dason you weren't being the push over I thought. We should ask quark to add a mom insult meter in addition to the thanks meter.

#### bryangoodrich

##### Probably A Mammal
I see Dason you weren't being the push over I thought. We should ask quark to add a mom insult meter in addition to the thanks meter.
lol I'll second that. I was surprised by his restraint. It is far superior to mine! So Hamid thought the cluster analysis built into SAS was just a file called cluster.sas?! So instead of asking, "how does SAS handle outliers in its PROC CLUSTER methods" he asks for a code file that doesn't even exist?