help using lag function

[wsu_wright]

I need to identify the period when a change occurs in the status of a
respondent.  Listed below is a sample data file

v1 = class (1 thru 6)
v2 = period (1 thru 72)

ID v1 v2
12  5  43
12  5  44
12  4  45
12  4  45
34  2  38
34  3  39
34  3  40
34  4  41
34  4  42

I'd like to create a flag to show when a change occurred from the
previous record-- v3.
ID v1 v2 v3
12  5  43 0
12  5  44 0
12  4  45 1
12  4  45 0
34  2  38 0
34  3  39 1
34  3  40 0
34  4  41 1
34  4  42 0

Thanks,

David

=====================
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

[Art Kendall]

see if syntax like that below is what you want.
Note that it will be very dependent on case order.



data list list/ID (f2) class(f1) period (f2) want (f1).
begin data
12� 5� 43 0
12� 5� 44 0
12� 4� 45 1
12� 4� 45 0
34� 2� 38 0
34� 3� 39 1
34� 3� 40 0
34� 4� 41 1
34� 4� 42 0
end data.
compute changed eq 0.
do if id eq lag(id).
compute changed =� class ne lag(class).
end if.
formats changed (f1).
LIST .

Art Kendall
Social Research Consultants

David Wright wrote:

I need to identify the period when a change occurs in the status of a
respondent.� Listed below is a sample data file

v1 = class (1 thru 6)
v2 = period (1 thru 72)

ID v1 v2
12� 5� 43
12� 5� 44
12� 4� 45
12� 4� 45
34� 2� 38
34� 3� 39
34� 3� 40
34� 4� 41
34� 4� 42

I'd like to create a flag to show when a change occurred from the
previous record-- v3.
ID v1 v2 v3
12� 5� 43 0
12� 5� 44 0
12� 4� 45 1
12� 4� 45 0
34� 2� 38 0
34� 3� 39 1
34� 3� 40 0
34� 4� 41 1
34� 4� 42 0

Thanks,

David

=====================
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

[Gary Oliver]

Hi All and SPSS

 

Here at University of Sydney, I have just received a message saying we are being upgraded from v15 to v16. Please can anyone explain to me why if v17 is out SPSS would not be providing codes for the less troublesome v17? Public info to the list or private information to my e-mail would be gratefully received

 

Warm regards/gary

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: Wednesday, February 11, 2009 2:06 AM
To: [hidden email]
Subject: Re: help using lag function

see if syntax like that below is what you want.
Note that it will be very dependent on case order.



data list list/ID (f2) class(f1) period (f2) want (f1).
begin data
12 5 43 0
12 5 44 0
12 4 45 1
12 4 45 0
34 2 38 0
34 3 39 1
34 3 40 0
34 4 41 1
34 4 42 0
end data.
compute changed eq 0.
do if id eq lag(id).
compute changed = class ne lag(class).
end if.
formats changed (f1).
LIST .

Art Kendall
Social Research Consultants

David Wright wrote:

I need to identify the period when a change occurs in the status of a
respondent. Listed below is a sample data file

v1 = class (1 thru 6)
v2 = period (1 thru 72)

ID v1 v2
12 5 43
12 5 44
12 4 45
12 4 45
34 2 38
34 3 39
34 3 40
34 4 41
34 4 42

I'd like to create a flag to show when a change occurred from the
previous record-- v3.
ID v1 v2 v3
12 5 43 0
12 5 44 0
12 4 45 1
12 4 45 0
34 2 38 0
34 3 39 1
34 3 40 0
34 4 41 1
34 4 42 0

Thanks,

David

=====================
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

[Ramzan Afzal-2]

Dear Members,

 

Our organization has recently upgarded SPSS and installed the latest version 17 base module. I observed some changes like the option of forecasting instead of time series. I would be grateful if you would be able to answer my query as mentioned below:

 

Do we need to get an add-on module SPSS Trends for v 17 or forecasting option is sufficient which is inlcuded in the base module of v17? 

 

 

 

Thanks to all

 

Regards

Ramzan

[SPSS Support]

Hi Ramzan,

 

Changes in this area began with Release 14:

 

In Release 14, SPSS added a new TSMODEL procedure (accessed in the menus via Analyze>Time Series>Create Models or Apply Models) and we're trying to guide people to the use of this procedure instead of the older AREG, ARIMA and EXSMOOTH procedures, which have now entered legacy status, which means that they will not have further work done on them.

 

If you have a new copy of SPSS 14 with a Trends license and are not recognized as having had the Trends module in prior releases, you will no longer see Autoregression, ARIMA and Exponential Smoothing choices on the Time Series menu and those procedures will not be available from command syntax. If you're recognized as upgrading to Release 14 you have two actual Trends licenses, and will be able to see all of those options and also to run those procedures through command syntax. In either case, the chapters for those three procedures have been removed from the SPSS Command Syntax Reference. If you have the proper license, you can get that information in the Command Syntax Reference under Help>Topics.

 

Release 15 functions in the same manner as Release 14.

 

In Release 16, the ARIMA, Autoregression and Exponential Smoothing dialogs have been removed for all users.

 

In Release 17, the Time Series grouping on the Analyze menu has been renamed Forecast.

 

If you have access to the Create Models and Apply Models options under Forecasting, then you do have a Trends license.

 

 

David Nichols

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ramzan Afzal
Sent: Wednesday, February 11, 2009 1:46 AM
To: [hidden email]
Subject: Re: [SPSSX-L] SPSS v16 not v17 university upgrade

Dear Members,

 

Our organization has recently upgarded SPSS and installed the latest version 17 base module. I observed some changes like the option of forecasting instead of time series. I would be grateful if you would be able to answer my query as mentioned below:

 

Do we need to get an add-on module SPSS Trends for v 17 or forecasting option is sufficient which is inlcuded in the base module of v17? 

 

 

 

Thanks to all

 

Regards

Ramzan