Creating a variable or function?

[David Marso-2]

VALUE LABELS AMBIGUOUS_POND_RMQ
"-1" 'All PREGYEARS missing'
"0" 'Matching Unique RMQ Preg Year'
"1" 'Multiple Possible RMQ Preg Years'.

Should be:

VALUE LABELS AMBIGUOUS_POND_RMQ
 -1 'All PREGYEARS missing'
 0 'Matching Unique RMQ Preg Year'
 1 'Multiple Possible RMQ Preg Years'.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[J.D. Haltigan]

Interesting. The double quotes around the values for the value labels has
always worked for me with no trouble.



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

I would expect it to give an error or warning but it works.
Don't plan to use it.  Something doesn't feel quite right .

DATA LIST FREE / x.
BEGIN DATA
1
END DATA.
VALUE LABELS
 x
 "1" "One".
FREQUENCIES x.



J.D. Haltigan wrote

> Interesting. The double quotes around the values for the value labels has
> always worked for me with no trouble.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Jon Peck]

Single and double quotes are perfectly legit anywhere in Statistics.  And inside either, you can use the other or double it.

On Mon, Apr 8, 2019 at 8:29 AM David Marso <[hidden email]> wrote:

I would expect it to give an error or warning but it works.
Don't plan to use it.  Something doesn't feel quite right .

DATA LIST FREE / x.
BEGIN DATA
1
END DATA.
VALUE LABELS
 x
 "1" "One".
FREQUENCIES x.



J.D. Haltigan wrote
> Interesting. The double quotes around the values for the value labels has
> always worked for me with no trouble.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

--
===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[David Marso]

These were for labeling a NUMERIC variable, hence my initial post.


Jon Peck wrote
> Single and double quotes are perfectly legit anywhere in Statistics.  And
> inside either, you can use the other or double it.
>
> On Mon, Apr 8, 2019 at 8:29 AM David Marso &lt;

> david.marso@

> &gt; wrote:
>
>> I would expect it to give an error or warning but it works.
>> Don't plan to use it.  Something doesn't feel quite right .
>>
>> DATA LIST FREE / x.
>> BEGIN DATA
>> 1
>> END DATA.
>> VALUE LABELS
>>  x
>>  "1" "One".
>> FREQUENCIES x.
>>
>>
>>
>> J.D. Haltigan wrote
>> > Interesting. The double quotes around the values for the value labels
>> has
>> > always worked for me with no trouble.
>> >
>> >
>> >
>> > --
>> > Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>> >
>> > =====================
>> > To manage your subscription to SPSSX-L, send a message to
>>
>> > LISTSERV@.UGA
>>
>> >  (not to SPSSX-L), with no body text except the
>> > command. To leave the list, send the command
>> > SIGNOFF SPSSX-L
>> > For a list of commands to manage subscriptions, send the command
>> > INFO REFCARD
>>
>>
>>
>>
>>
>> -----
>> Please reply to the list and not to my personal email.
>> Those desiring my consulting or training services please feel free to
>> email me.
>> ---
>> "Nolite dare sanctum canibus neque mittatis margaritas vestras ante
>> porcos
>> ne forte conculcent eas pedibus suis."
>> Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff
>> in
>> abyssum?"
>> --
>> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>>
>> =====================
>> To manage your subscription to SPSSX-L, send a message to
>>

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
>> command. To leave the list, send the command
>> SIGNOFF SPSSX-L
>> For a list of commands to manage subscriptions, send the command
>> INFO REFCARD
>>
> --
> Jon K Peck

> jkpeck@

>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[J.D. Haltigan]

Wanted to follow-up with an elementary query on this thread rather than
clutter the board with a useless one.

I am working with a raw data set that is a behemoth and creating summary
variables for a subsequent analyses. Part of my strategy is to
create/compute relevant summary variables and place them at the end of the
working raw dataset. I will then DROP variables when ready. For some
variables I just want to essentially copy them to the end (slightly
different variant of the variable name, of course).

Right now I just use COMPUTE V1_VARIANT=VARIANT.
EXE.

This is fine and serves my purposes, but I was wondering if there was
something more relevant/efficient/informative.

Thanks for any suggestions.



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Rich Ulrich]

Well, I would not DROP variables. I keep my "real analysis" files as

permanent documentation. For a computation that you sometimes

want/ sometimes don't want, you might put the syntax into a file to

be INSERTed.  Or the reusable summary scores could be saved in a

smaller file with just ID, Scores.

Is your dataset behemoth because of number of variables?  I grew up

with slow computers and slow processing, so I learned to strip my "big"

file down to only the variables that I needed for a set of analyses. I

imagine that also helped to prevent mistakes of using the wrong

variable with a too-similar name, or failing to include part of the

essential set. And that file would be saved as permanent documentation.

--

Rich Ulrich

---

From: SPSSX(r) Discussion <[hidden email]> on behalf of J.D. Haltigan <[hidden email]>
Sent: Sunday, April 14, 2019 9:39 PM
To: [hidden email]
Subject: Re: Creating a variable or function?

 

Wanted to follow-up with an elementary query on this thread rather than
clutter the board with a useless one.

I am working with a raw data set that is a behemoth and creating summary
variables for a subsequent analyses. Part of my strategy is to
create/compute relevant summary variables and place them at the end of the
working raw dataset. I will then DROP variables when ready. For some
variables I just want to essentially copy them to the end (slightly
different variant of the variable name, of course).

Right now I just use COMPUTE V1_VARIANT=VARIANT.
EXE.

This is fine and serves my purposes, but I was wondering if there was
something more relevant/efficient/informative.

Thanks for any suggestions.



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[J.D. Haltigan]

I did not mean to imply I would not save the original file, Rich. Indeed, I
would drop and 'pair down' my analytic file to only those variables I need
for subsequent analyses saving as a new file etc. Of course all this gets
documented both in the syntax file as well as in a separate document.



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso-2]

RECODE VARIANT (COPY) INTO V1_VARIANT.

Perhaps.  I have not done any benchmarks.
At any rate get rid of the EXE if you have a behemoth file.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

Should be

RECODE VARIANT (ELSE=COPY) INTO V1_VARIANT.

Slightly slower than COMPUTE.

MATRIX.
SAVE MAKE(30000000,1,1)/ OUTFILE */VARIABLES test.
END MATRIX.
COMPUTE test=TRUNC(RV.UNIFORM(1,6)).
EXECUTE.
DATASET NAME rawtest.

MATRIX.
SAVE MAKE(1,1,1)/ OUTFILE */VARIABLES bench.
END MATRIX.
DATASET NAME bench.

COMPUTE T0=$TIME.
DATASET ACTIVATE rawtest.
COMPUTE test_new=test.
EXECUTE.
DATASET ACTIVATE bench.
COMPUTE T_elapsed=$TIME-t0.
PRINT /T_elapsed.
EXECUTE.

COMPUTE T0=$TIME.
DATASET ACTIVATE rawtest.
RECODE test (ELSE=COPY) INTO test_new.
EXECUTE.
DATASET ACTIVATE bench.
COMPUTE T_elapsed=$TIME-t0.
PRINT /T_elapsed.
EXECUTE.



David Marso-2 wrote
> RECODE VARIANT (COPY) INTO V1_VARIANT.
>
> Perhaps.  I have not done any benchmarks.
> At any rate get rid of the EXE if you have a behemoth file.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[J.D. Haltigan]

Based on this would seem the COMPUTE route is marginally a more efficient way
to do it. Would you agree?

Also re: my prior post I want to correct to 'pare down' NOT pair down as
previously stated.



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

True.  It is also more concise unless you are copying many variables.
OTOH: DO REPEAT will help with that.

J.D. Haltigan wrote

> Based on this would seem the COMPUTE route is marginally a more efficient
> way
> to do it. Would you agree?
>
> Also re: my prior post I want to correct to 'pare down' NOT pair down as
> previously stated.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

>  (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD





-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD