|
Hi,
I've just signed up for this list and I'm not even sure if it's still running since the "last updated date" on the bottom of the page was 2002. My problem is one that I feel should be simple to solve, but not having much spss syntax background, I'm not sure how to make it work. I have a list of accounts, and various prices associated with them. In java, if I wanted to apply something to each group of accounts, I would use a loop. (for i <=n total = cost[i] + total ) or something along those lines. Essentially, I am trying to get the program to put the total of the whole account on every line for that account. I have tried the casestovars command, but it is not quite the format I am looking for. Any help would be greatly appreciated. Thanks a lot Tiffany |
|
At 01:33 PM 7/5/2007, Tiffany Swinimer wrote:
>I've just signed up for this list Well, welcome, and we're glad to have you. >I'm not even sure if it's still running since the "last updated date" >on the bottom of the page was 2002. Well, I think we're still running. Yours is, indeed, only the 7th posting today, but it *is* the day after a holiday. >I have a list of accounts, and various prices associated with them. OK, to start: It looks like an account has a number of SPSS records ('cases'); every record has a price; and >I am trying to get the program to put the total of the whole account >on every line for that account. You have to have a variable (or a set of variables) that uniquely identify the account; I'll call it 'Account#'. To add all the values of Price for the account, and put the total in every record for that account, then (not tested), AGGREGATE OUTFILE=* MODE=ADDVARIABLES /BREAK=Account# /TotPrice 'Total of all prices for account'=SUM(Price). If I've misunderstood, or you have an early SPSS version without 'MODE=ADDVARIABLES', or anything else doesn't work, post again. -Welcome and good luck, Richard Ristow |
|
In reply to this post by Tiffany Swinimer
Tiffany,
I not quite sure what you want from your description but is sounds like this example might be correct. Rec price total 1 5.05 5.05 2 1.23 6.28 3 3.22 9.50 4 4.00 13.50 5 9.95 15.00 So that for each record, total is the value on the previous record plus the value of price on the current record. Could this be correct? A really big problem is whether a file contains one account or multiple accounts. I'll assume the former because it is easier. Do if ($casenum eq 1). + compute total=price. Else. + compute total=price+lag(total). End if. |
|
In reply to this post by Tiffany Swinimer
Tiffany, if your data is in the format:
Account Cost1 Cost2 Cost3 A 90 23 23 B 23 12 45 ... Then the command "COMPUTE Total = SUM(Cost1 TO Cost3)" will create: Account Cost1 Cost2 Cost3 Total A 90 23 23 136 B 23 12 45 80 ... Dan Williams Forecasting, Research and Analysis Office Department of Human Services State of Oregon, USA 503 947 5354 >>> "Tiffany Swinimer" <[hidden email]> 7/5/2007 10:33:47 AM >>> Hi, I've just signed up for this list and I'm not even sure if it's still running since the "last updated date" on the bottom of the page was 2002. My problem is one that I feel should be simple to solve, but not having much spss syntax background, I'm not sure how to make it work. I have a list of accounts, and various prices associated with them. In java, if I wanted to apply something to each group of accounts, I would use a loop. (for i <=n total = cost[i] + total ) or something along those lines. Essentially, I am trying to get the program to put the total of the whole account on every line for that account. I have tried the casestovars command, but it is not quite the format I am looking for. Any help would be greatly appreciated. Thanks a lot Tiffany |
|
In reply to this post by Tiffany Swinimer
Is your goal to produce some sort of report or do you want data aggregated
for further analysis? SPSS has a number of tabulation and reporting procedures which can be used to sum some variable x across instances of another variable. Go to the Help index to find information on procedures such as REPORT, SUMMARIZE, TABLES, and CTABLES. If you want to aggregate data for further analysis, look to the AGGREGATE to produce summarized data. TABLES and CTABLES are not part of base SPSS but provide many options for creating a customized report. You'll find that many posters on this list prefer to use SPSS syntax but don't overlook the dialog boxes from the menus. You'll find the reporting procedures on the Analysis menu. Sometimes you'll have to write your own syntax to handle special situations both in desired output and unusual data structures. From the basic description, this doesn't sound like one of them. You'll find more direct and helpful responses if you describe your data--what defines a case (row) of data and what variables are defined for that case. Finally, let us know what release of SPSS you are running because solutions will differ by release. -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Tiffany Swinimer Sent: Thursday, July 05, 2007 11:34 AM To: [hidden email] Subject: Loops Hi, I've just signed up for this list and I'm not even sure if it's still running since the "last updated date" on the bottom of the page was 2002. My problem is one that I feel should be simple to solve, but not having much spss syntax background, I'm not sure how to make it work. I have a list of accounts, and various prices associated with them. In java, if I wanted to apply something to each group of accounts, I would use a loop. (for i <=n total = cost[i] + total ) or something along those lines. Essentially, I am trying to get the program to put the total of the whole account on every line for that account. I have tried the casestovars command, but it is not quite the format I am looking for. Any help would be greatly appreciated. Thanks a lot Tiffany |
| Free forum by Nabble | Edit this page |
