* Dear SPSS friends, I'm doing several hundred regression analyses with different versions of independend variables and want to export respective model information into a file; the information I need is R, the regression coefficients, and the coefficient significances; As a simplified example I've tried this syntax but multiple R is unfortunately not exported. DATA LIST /dv 1 iv1 2 iv2 3 iv3 4 iv4 5. BEGIN DATA 5255 2341 4234 3433 2142 END DATA. REGRESSION /MISSING LISTWISE /STATISTICS COEFF R /CRITERIA=PIN(.05) POUT(.10) /NOORIGIN /DEPENDENT dv /METHOD=ENTER iv1 iv2 iv3 /OUTFILE CORB ('ra_results.sav') . * Is there a convenient way to get names of independend variables, R, the regression coefficients, and the coefficient significances into a file so that a quick comparison between models is possible? A text file would be fine as well; Thanks for any opinions, Mario.
Mario Giesel
Munich, Germany |
Administrator
|
Mario,
I believe you will want to delve into that wonderful thing called OMS! HTH, David
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?" |
Administrator
|
Indeed! I was about to suggest the same thing. Something like this, perhaps...
NEW FILE. DATASET CLOSE all. DATA LIST /dv 1 iv1 2 iv2 3 iv3 4 iv4 5. BEGIN DATA 5255 2341 4234 3433 2142 END DATA. DATASET NAME raw. * OMS. DATASET DECLARE Coeff. DATASET DECLARE Summary. OMS /SELECT TABLES /IF COMMANDS=['Regression'] SUBTYPES=['Coefficients'] /DESTINATION FORMAT=SAV NUMBERED=TableNumber_ OUTFILE='Coeff'. OMS /SELECT TABLES /IF COMMANDS=['Regression'] SUBTYPES=[' Model Summary'] /DESTINATION FORMAT=SAV NUMBERED=TableNumber_ OUTFILE='Summary'. REGRESSION /MISSING LISTWISE /STATISTICS COEFF R /CRITERIA=PIN(.05) POUT(.10) /NOORIGIN /DEPENDENT dv /METHOD=ENTER iv1 iv2 iv3 . OMSEND. DATASET ACTIVATE Coeff. DELETE VARIABLES Command_ TO Var1. DATASET ACTIVATE Summary. DELETE VARIABLES Command_ TO Var1. MATCH FILES FILE = 'Coeff' / TABLE = 'Summary' / BY TableNumber_. EXECUTE. DATASET NAME ra_results. DATASET CLOSE Coeff. DATASET CLOSE Summary. DATASET ACTIVATE ra_results. LIST.
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
Administrator
|
The whole 'doing several hundred regressions...' idea gives me pause ;-(
There is probably some reasonable theory somewhere and some sort of APSS search is NOT likely to magically 'discover' it! I can't imagine how you are feeding this beast (well I can, but really don't want to think about it too hard). However OMS will support multiple REGRESSION commands between the OMS and OMSEND tags and serialize them into a stacked dataset indexed by TableNumber_ (in Bruce's example). OTOH: The /OUTFILE will *NOT* (it will only grab the last set of results). So, jump in and see what it will do for you -)
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?" |
Administrator
|
Yeah, it does sound like a fishing expedition, doesn't it!
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
Administrator
|
There's a pile of steaming sashimi in that mess somewhere ;-)
Serve it up with a giant pinch of wasabi and copious amounts of soy sauce! --
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?" |
Thanks for your comments, guys. Hope this did not spoil your appetite ;-) I tried Bruces' approach and am quite happy with it. Used OMS only for xlsx export so far where deletion of information is not so easy. Export to sav datasets works fine. Thanks & bye, Mario David Marso <[hidden email]> schrieb am 17:39 Mittwoch, 12.November 2014: There's a pile of steaming sashimi in that mess somewhere ;-) Serve it up with a giant pinch of wasabi and copious amounts of soy sauce! -- Bruce Weaver wrote > Yeah, it does sound like a fishing expedition, doesn't it! > David Marso wrote >> The whole 'doing several hundred regressions...' idea gives me pause ;-( >> There is probably some reasonable theory somewhere and some sort of APSS >> search is NOT likely to magically 'discover' it! >> I can't imagine how you are feeding this beast (well I can, but really >> don't want to think about it too hard). >> However OMS will support multiple REGRESSION commands between the OMS and >> OMSEND tags and serialize them into a stacked dataset indexed by >> TableNumber_ (in Bruce's example). >> OTOH: The /OUTFILE will *NOT* (it will only grab the last set of >> results). >> So, jump in and see what it will do for you -) >> Bruce Weaver wrote >>> Indeed! I was about to suggest the same thing. Something like this, >>> perhaps... >>> >>> NEW FILE. >>> DATASET CLOSE all. >>> DATA LIST /dv 1 iv1 2 iv2 3 iv3 4 iv4 5. >>> BEGIN DATA >>> 5255 >>> 2341 >>> 4234 >>> 3433 >>> 2142 >>> END DATA. >>> DATASET NAME raw. >>> * OMS. >>> DATASET DECLARE Coeff. >>> DATASET DECLARE Summary. >>> >>> OMS >>> /SELECT TABLES >>> /IF COMMANDS=['Regression'] SUBTYPES=['Coefficients'] >>> /DESTINATION FORMAT=SAV NUMBERED=TableNumber_ >>> OUTFILE='Coeff'. >>> OMS >>> /SELECT TABLES >>> /IF COMMANDS=['Regression'] SUBTYPES=[' Model Summary'] >>> /DESTINATION FORMAT=SAV NUMBERED=TableNumber_ >>> OUTFILE='Summary'. >>> >>> REGRESSION >>> /MISSING LISTWISE >>> /STATISTICS COEFF R >>> /CRITERIA=PIN(.05) POUT(.10) >>> /NOORIGIN >>> /DEPENDENT dv >>> /METHOD=ENTER iv1 iv2 iv3 >>> . >>> OMSEND. >>> >>> DATASET ACTIVATE Coeff. >>> DELETE VARIABLES Command_ TO Var1. >>> DATASET ACTIVATE Summary. >>> DELETE VARIABLES Command_ TO Var1. >>> >>> MATCH FILES >>> FILE = 'Coeff' / >>> TABLE = 'Summary' / >>> BY TableNumber_. >>> EXECUTE. >>> DATASET NAME ra_results. >>> DATASET CLOSE Coeff. >>> DATASET CLOSE Summary. >>> DATASET ACTIVATE ra_results. >>> LIST. >>> >>> David Marso wrote >>>> Mario, >>>> I believe you will want to delve into that wonderful thing called >>>> OMS! >>>> HTH, David >>>> Mario Giesel wrote >>>>> * Dear SPSS friends, I'm doing several hundred regression analyses >>>>> with different versions of independend variables >>>>> and want to export respective model information into a file; >>>>> the information I need is R, the regression coefficients, and the >>>>> coefficient significances; >>>>> As a simplified example I've tried this syntax but multiple R is >>>>> unfortunately not exported. >>>>> >>>>> DATA LIST /dv 1 iv1 2 iv2 3 iv3 4 iv4 5. >>>>> BEGIN DATA >>>>> 5255 >>>>> 2341 >>>>> 4234 >>>>> 3433 >>>>> 2142 >>>>> END DATA. >>>>> >>>>> REGRESSION >>>>> /MISSING LISTWISE >>>>> /STATISTICS COEFF R >>>>> /CRITERIA=PIN(.05) POUT(.10) >>>>> /NOORIGIN >>>>> /DEPENDENT dv >>>>> /METHOD=ENTER iv1 iv2 iv3 >>>>> /OUTFILE CORB ('ra_results.sav') . >>>>> >>>>> * Is there a convenient way to get names of independend variables, R, >>>>> the regression coefficients, and the coefficient significances >>>>> into a file so that a quick comparison between models is possible? A >>>>> text file would be fine as well; >>>>> Thanks for any opinions, >>>>> Mario. >>>>> >>>>> ===================== >>>>> 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 ----- 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?" -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Regression-Results-Export-tp5727880p5727890.html Sent from the SPSSX Discussion mailing list archive at 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
Mario Giesel
Munich, Germany |
Free forum by Nabble | Edit this page |