RE: restructuring-many records into one

[prasad]

CONTENTS DELETED

The author has deleted this message.

[David Marso]

What should the final file look like?
If you want all diagXX to be in a single line with mastend as the case I would suggest going from multivariable to single variable (VARSTOCASES) followed by CASESTOVARS.  Otherwise you have additional pain.
What do you wish to determine from the final restructured data?
Better to have an upload of the actual anonymized data than a screenshot.

prasad wrote

Hi David,

Apologies for not furnishing the full information.

1.      There are diag20 in the file

2.      I built and index with a view to use it, but had trouble populating the vector.

3.      CASESTOVARS will be tried.

I have attached a screenshot of the data file (var are shown only till 13).

Your input will be much appreciated.

Thanks
Prasad

From: David Marso [via SPSSX Discussion] [mailto:[hidden email]]
Sent: Friday, 12 October 2012 10:33 a.m.
To: Prasad Nishtala
Subject: Re: restructring-many records into one

What do you expect this to do?
Maybe you should look at CASESTOVARS instead of trying to run code which it appears you do not understand.
eg VECTOR X=diag01 to diag20.
You only have up to diag06 in your data file
and you build an index? tasknum
but don't use it to populate the vector.
--
prasad wrote
How to convert this:
    mast_enc     ENDTYPE   FUND diag01     diag0 diag03 diag04 diag05 diag06
..0.GVcV2Xk     DR       N          E1120  N1891  D638    E1170  H350    Z8643
..0.GVcV2Xk     DR       N          E1120  N1891  D638    E1170  I500     H350    J22
..0.GVcV2Xk     DR       N          K529


To
..0.GVcV2Xk     DR       N          E1120  N1891  D638    E1170  H350    Z8643  1500 J22 K529

i tried this, but it only pickups the first record.

COMPUTE tasknum=1.
DO IF ($CASENUM NE 1).
IF (MAST_ENC EQ LAG(MAST_ENC)) tasknum = LAG(tasknum)+1.
END IF.
VECTOR X=diag01 to diag20.
AGGREGATE
  /OUTFILE=*
  /BREAK=mast_enc
  /diag01 to diag20= max (diag01 to diag 20).
EXECUTE.

your help is invaulable.

prasad
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.

________________________________
If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/restructring-many-records-into-one-tp5715605p5715606.html
To unsubscribe from restructring-many records into one, click here<http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=5715605&code=cHJhc2FkLm5pc2h0YWxhQG90YWdvLmFjLm56fDU3MTU2MDV8LTgyMTM2Mjc=>.
NAML<http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>


datafile.png (225K) <http://spssx-discussion.1045642.n5.nabble.com/attachment/5715607/0/datafile.png>

[prasad]

Thanks David,
Agree, my intention was to have all diag in 1 line with mastend as case 1.
Will VARSTOCASES AND CASESTOVARS work when you have more than 400,000 records.
Point noted. The data is anonymised, but agree supplying a screenshot is not a good idea.
I will be linking it to the prescription data.
Prasad

[David Marso]

No problemo!
Consider:
input program.
+  numeric id rec (n6.0).
+  string diag01 to diag20 (a8).
+  vector diag=diag01 to diag20.
+  loop id=1 to 100000.
+    compute #nrec=trunc(uniform(20)).
+    loop rec=1 to #nrec.
+      compute #ndiag=trunc(uniform(20)).
+      loop #d=1 to #ndiag.
+        compute #x= trunc(uniform(20)).
+        compute diag(#d)=string(#x,N5).
+      end loop.
+      leave id.
+      end case.
+    end loop.
+  end loop.
+  end file.
end input program.
exe.

VARSTOCASES  /MAKE trans1 FROM diag01 to diag20 /KEEP =  id  /NULL = DROP.
CASESTOVARS /ID = id /GROUPBY = VARIABLE .

prasad wrote

Thanks David,
Agree, my intention was to have all diag in 1 line with mastend as case 1.
Will VARSTOCASES AND CASESTOVARS work when you have more than 400,000 records.
Point noted. The data is anonymised, but agree supplying a screenshot is not a good idea.
I will be linking it to the prescription data.
Prasad

[Veena Nambiar]

I am currently out of the office and will return your message upon my return on Monday, October 15th. Thanks.

[prasad]

Gracias. I will run the syntax and let you know.

Cheers Prasad

 

---

From: David Marso [via SPSSX Discussion] [ml-node+[hidden email]]
Sent: Friday, 12 October 2012 1:43 PM
To: Prasad Nishtala
Subject: RE: restructuring-many records into one

No problemo!
Consider:
input program.
+  numeric id rec (n6.0).
+  string diag01 to diag20 (a8).
+  vector diag=diag01 to diag20.
+  loop id=1 to 100000.
+    compute #nrec=trunc(uniform(20)).
+    loop rec=1 to #nrec.
+      compute #ndiag=trunc(uniform(20)).
+      loop #d=1 to #ndiag.
+        compute #x= trunc(uniform(20)).
+        compute diag(#d)=string(#x,N5).
+      end loop.
+      leave id.
+      end case.
+    end loop.
+  end loop.
+  end file.
end input program.
exe.

VARSTOCASES  /MAKE trans1 FROM diag01 to diag20 /KEEP =  id  /NULL = DROP.
CASESTOVARS /ID = id /GROUPBY = VARIABLE .

prasad wrote

Thanks David,
Agree, my intention was to have all diag in 1 line with mastend as case 1.
Will VARSTOCASES AND CASESTOVARS work when you have more than 400,000 records.
Point noted. The data is anonymised, but agree supplying a screenshot is not a good idea.
I will be linking it to the prescription data.
Prasad

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/RE-restructuring-many-records-into-one-tp5715607p5715612.html

To unsubscribe from RE: restructuring-many records into one, click here.
NAML

[prasad]

thanks david,

varstocases worked!!

---

From: David Marso [via SPSSX Discussion] [ml-node+[hidden email]]
Sent: Friday, 12 October 2012 1:43 PM
To: Prasad Nishtala
Subject: RE: restructuring-many records into one

No problemo!
Consider:
input program.
+  numeric id rec (n6.0).
+  string diag01 to diag20 (a8).
+  vector diag=diag01 to diag20.
+  loop id=1 to 100000.
+    compute #nrec=trunc(uniform(20)).
+    loop rec=1 to #nrec.
+      compute #ndiag=trunc(uniform(20)).
+      loop #d=1 to #ndiag.
+        compute #x= trunc(uniform(20)).
+        compute diag(#d)=string(#x,N5).
+      end loop.
+      leave id.
+      end case.
+    end loop.
+  end loop.
+  end file.
end input program.
exe.

VARSTOCASES  /MAKE trans1 FROM diag01 to diag20 /KEEP =  id  /NULL = DROP.
CASESTOVARS /ID = id /GROUPBY = VARIABLE .

prasad wrote

Thanks David,
Agree, my intention was to have all diag in 1 line with mastend as case 1.
Will VARSTOCASES AND CASESTOVARS work when you have more than 400,000 records.
Point noted. The data is anonymised, but agree supplying a screenshot is not a good idea.
I will be linking it to the prescription data.
Prasad

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/RE-restructuring-many-records-into-one-tp5715607p5715612.html

To unsubscribe from RE: restructuring-many records into one, click here.
NAML