Logistic Regression with conditions
Locked
Logistic Regression with conditions
 
|
Hi all,
I am attempting to carry out a logistic regression on horse race data as follows. Dependent: win (1= horse wins, 0 = horse loses). Status variable: date of horse race Stratifying variable: race number (number of horses in a race varies). Up to 243 independent variables. I have attempted to use the COX regression syntax posted by Marta Garcia-Granero. The issue I have is that not only does the number of horse in each race change, so does the actual horses in it. Any ideas? Max ===================== 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 |
Re: Logistic Regression with conditions
|
|
Chipulu M. WROTE:
> > I am attempting to carry out a logistic regression on horse race data as follows. > > Dependent: win (1= horse wins, 0 = horse loses). > > Status variable: date of horse race > > Stratifying variable: race number (number of horses in a race varies). > > Up to 243 independent variables. > > I have attempted to use the COX regression syntax posted by Marta Garcia-Granero. The issue I have is that not only does the number of horse in each race change, so does the actual horses in it. Hi As long as only one horse wins (as is to be expected), the number of losers can vary 8from 1 to n) without any problem Race Win PTime 1 1 1 1 0 2 1 0 2 . . . . 1 0 2 2 1 1 2 0 2 . . 2 0 2 But the status variable is what you call the dependent (win=1/loose=0), not the horse race date (this last variable does not play a role in conditional logistic regression). The role of time is played by a "pseudo-time" variable where ptime=2 if horse looses, and ptime=1 if horse wins. HTH, Marta García-Granero -- For miscellaneous SPSS related statistical stuff, visit: http://gjyp.nl/marta/ ===================== 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 |
|
|
Hello to all:
I'm sure for many of you, question below sounds foolish, I extremely apologize for this. I'm running crosstabs (2x2) and unable to getting risk estimate because 40 people among 450 is with missing values. What is that I'm not aware? Is it truly due to the missing data? Here is the notice I'm having: Risk estimate statistics cannot be computed. They are only computed for a 2*2 table without empty cells. Many thanks for your help! Emel -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Marta García-Granero Sent: Wednesday, April 29, 2009 6:24 AM To: [hidden email] Subject: Re: Logistic Regression with conditions Chipulu M. WROTE: > > I am attempting to carry out a logistic regression on horse race data as follows. > > Dependent: win (1= horse wins, 0 = horse loses). > > Status variable: date of horse race > > Stratifying variable: race number (number of horses in a race varies). > > Up to 243 independent variables. > > I have attempted to use the COX regression syntax posted by Marta each race change, so does the actual horses in it. Hi As long as only one horse wins (as is to be expected), the number of losers can vary 8from 1 to n) without any problem Race Win PTime 1 1 1 1 0 2 1 0 2 . . . . 1 0 2 2 1 1 2 0 2 . . 2 0 2 But the status variable is what you call the dependent (win=1/loose=0), not the horse race date (this last variable does not play a role in conditional logistic regression). The role of time is played by a "pseudo-time" variable where ptime=2 if horse looses, and ptime=1 if horse wins. HTH, Marta García-Granero -- For miscellaneous SPSS related statistical stuff, visit: http://gjyp.nl/marta/ ===================== 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 The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail. ===================== 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 |
|
|
Note that the error message says you have no cases in one of your cells. It
could be due to missing data or it could be due that the cell is systematically empty as might be seen with a risk analysis of gender and pregnancy--males can't be pregnant. Next time you send a question to this list would you please trim off the contents of the some other message having nothing to do with your query. -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ergul, Emel A. Sent: Wednesday, April 29, 2009 9:09 AM To: [hidden email] Subject: Cross tabs/ risk assessment Hello to all: I'm sure for many of you, question below sounds foolish, I extremely apologize for this. I'm running crosstabs (2x2) and unable to getting risk estimate because 40 people among 450 is with missing values. What is that I'm not aware? Is it truly due to the missing data? Here is the notice I'm having: Risk estimate statistics cannot be computed. They are only computed for a 2*2 table without empty cells. Many thanks for your help! Emel <snip/> ===================== 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 |
Re: Cross tabs/ risk assessment
|
|
Thanks very much ViAnn.
emel -----Original Message----- From: ViAnn Beadle [mailto:[hidden email]] Sent: Wednesday, April 29, 2009 11:29 AM To: Ergul, Emel A.; [hidden email] Subject: RE: Cross tabs/ risk assessment Note that the error message says you have no cases in one of your cells. It could be due to missing data or it could be due that the cell is systematically empty as might be seen with a risk analysis of gender and pregnancy--males can't be pregnant. Next time you send a question to this list would you please trim off the contents of the some other message having nothing to do with your query. The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail. ===================== 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 |
|
|
In reply to this post by ViAnn Beadle
Hello all,
I am doing some time series forecasting using ARIMA in V17. I used the menu to build my model, but then pasted the syntax. The model I ran had a AR value of 2, but this is what was pasted in the syntax window: /ARIMA AR=[2,1] DIFF=0 MA=[1] ARSEASONAL=[0] DIFFSEASONAL=0 MASEASONAL=[0] TRANSFORM=SQRT CONSTANT=YES Can anyone help me understand what the AR = [2,1] parameter means and how it is different from AR = [2]? Thanks. Rodrigo Guerrero Director of Marketing Research and Analysis The SCOOTER Store Office: 830-627-4317 Mobile: 830-832-8490 We provide FREEDOM and INDEPENDENCE to people with limited mobility The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 |
|
|
HELP on the TSMODEL syntax says that this specifies 1st and 2nd lags. HELP
on the Time Series Modeler says the following: " All positive lower orders will be included in the model. For example, if you specify 2, the model includes orders 2 and 1." -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo Sent: Wednesday, April 29, 2009 9:48 AM To: [hidden email] Subject: ARIMA Modeling Hello all, I am doing some time series forecasting using ARIMA in V17. I used the menu to build my model, but then pasted the syntax. The model I ran had a AR value of 2, but this is what was pasted in the syntax window: /ARIMA AR=[2,1] DIFF=0 MA=[1] ARSEASONAL=[0] DIFFSEASONAL=0 MASEASONAL=[0] TRANSFORM=SQRT CONSTANT=YES Can anyone help me understand what the AR = [2,1] parameter means and how it is different from AR = [2]? Thanks. Rodrigo Guerrero Director of Marketing Research and Analysis The SCOOTER Store Office: 830-627-4317 Mobile: 830-832-8490 We provide FREEDOM and INDEPENDENCE to people with limited mobility The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 |
|
|
Thank you. I have some models that have the AR = [2,1] combination and
some that have the AR = [2]. It does not seem to be consistent and I cannot figure out the difference. The models are predicting different data series, but are identical in the way they got set up in the dialog. RG Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle Sent: Wednesday, April 29, 2009 11:58 AM To: [hidden email] Subject: Re: ARIMA Modeling HELP on the TSMODEL syntax says that this specifies 1st and 2nd lags. HELP on the Time Series Modeler says the following: " All positive lower orders will be included in the model. For example, if you specify 2, the model includes orders 2 and 1." -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo Sent: Wednesday, April 29, 2009 9:48 AM To: [hidden email] Subject: ARIMA Modeling Hello all, I am doing some time series forecasting using ARIMA in V17. I used the menu to build my model, but then pasted the syntax. The model I ran had a AR value of 2, but this is what was pasted in the syntax window: /ARIMA AR=[2,1] DIFF=0 MA=[1] ARSEASONAL=[0] DIFFSEASONAL=0 MASEASONAL=[0] TRANSFORM=SQRT CONSTANT=YES Can anyone help me understand what the AR = [2,1] parameter means and how it is different from AR = [2]? Thanks. Rodrigo Guerrero Director of Marketing Research and Analysis The SCOOTER Store Office: 830-627-4317 Mobile: 830-832-8490 We provide FREEDOM and INDEPENDENCE to people with limited mobility The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 |
|
|
Based upon what Help says--this smells like a syntax generation bug. I'm not
very knowledgeable about ARIMA and can't get you much advice. Syntax generation bugs are always easy to get around by pasting the syntax and editing it to get the model you want. -----Original Message----- From: Guerrero, Rodrigo [mailto:[hidden email]] Sent: Wednesday, April 29, 2009 3:25 PM To: ViAnn Beadle; [hidden email] Subject: RE: Re: ARIMA Modeling Thank you. I have some models that have the AR = [2,1] combination and some that have the AR = [2]. It does not seem to be consistent and I cannot figure out the difference. The models are predicting different data series, but are identical in the way they got set up in the dialog. RG Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle Sent: Wednesday, April 29, 2009 11:58 AM To: [hidden email] Subject: Re: ARIMA Modeling HELP on the TSMODEL syntax says that this specifies 1st and 2nd lags. HELP on the Time Series Modeler says the following: " All positive lower orders will be included in the model. For example, if you specify 2, the model includes orders 2 and 1." -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo Sent: Wednesday, April 29, 2009 9:48 AM To: [hidden email] Subject: ARIMA Modeling Hello all, I am doing some time series forecasting using ARIMA in V17. I used the menu to build my model, but then pasted the syntax. The model I ran had a AR value of 2, but this is what was pasted in the syntax window: /ARIMA AR=[2,1] DIFF=0 MA=[1] ARSEASONAL=[0] DIFFSEASONAL=0 MASEASONAL=[0] TRANSFORM=SQRT CONSTANT=YES Can anyone help me understand what the AR = [2,1] parameter means and how it is different from AR = [2]? Thanks. Rodrigo Guerrero Director of Marketing Research and Analysis The SCOOTER Store Office: 830-627-4317 Mobile: 830-832-8490 We provide FREEDOM and INDEPENDENCE to people with limited mobility The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 |
|
|
I could be wrong, but it seems that [2] will run an ARIMA model with autoregressive parameter of order 2 only. The second model as ViAnn inferred should have both parameters of order 1 and 2 in the model. If you ran the model you should have more parameters in the second one. At least that is how I run mine in SAS.
-----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle Sent: Wednesday, April 29, 2009 3:00 PM To: [hidden email] Subject: Re: ARIMA Modeling Based upon what Help says--this smells like a syntax generation bug. I'm not very knowledgeable about ARIMA and can't get you much advice. Syntax generation bugs are always easy to get around by pasting the syntax and editing it to get the model you want. -----Original Message----- From: Guerrero, Rodrigo [mailto:[hidden email]] Sent: Wednesday, April 29, 2009 3:25 PM To: ViAnn Beadle; [hidden email] Subject: RE: Re: ARIMA Modeling Thank you. I have some models that have the AR = [2,1] combination and some that have the AR = [2]. It does not seem to be consistent and I cannot figure out the difference. The models are predicting different data series, but are identical in the way they got set up in the dialog. RG Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle Sent: Wednesday, April 29, 2009 11:58 AM To: [hidden email] Subject: Re: ARIMA Modeling HELP on the TSMODEL syntax says that this specifies 1st and 2nd lags. HELP on the Time Series Modeler says the following: " All positive lower orders will be included in the model. For example, if you specify 2, the model includes orders 2 and 1." -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo Sent: Wednesday, April 29, 2009 9:48 AM To: [hidden email] Subject: ARIMA Modeling Hello all, I am doing some time series forecasting using ARIMA in V17. I used the menu to build my model, but then pasted the syntax. The model I ran had a AR value of 2, but this is what was pasted in the syntax window: /ARIMA AR=[2,1] DIFF=0 MA=[1] ARSEASONAL=[0] DIFFSEASONAL=0 MASEASONAL=[0] TRANSFORM=SQRT CONSTANT=YES Can anyone help me understand what the AR = [2,1] parameter means and how it is different from AR = [2]? Thanks. Rodrigo Guerrero Director of Marketing Research and Analysis The SCOOTER Store Office: 830-627-4317 Mobile: 830-832-8490 We provide FREEDOM and INDEPENDENCE to people with limited mobility The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 NOTICE: This e-mail (and any attachments) may contain PRIVILEGED OR CONFIDENTIAL information and is intended only for the use of the specific individual(s) to whom it is addressed. It may contain information that is privileged and confidential under state and federal law. This information may be used or disclosed only in accordance with law, and you may be subject to penalties under law for improper use or further disclosure of the information in this e-mail and its attachments. If you have received this e-mail in error, please immediately notify the person named above by reply e-mail, and then delete the original e-mail. Thank you. ===================== 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 |
|
|
In reply to this post by ViAnn Beadle
Thank you. That is just the way the cookie crumbles, I guess.
RG Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317 -----Original Message----- From: ViAnn Beadle [mailto:[hidden email]] Sent: Wednesday, April 29, 2009 5:00 PM To: Guerrero, Rodrigo Cc: [hidden email] Subject: RE: Re: ARIMA Modeling Based upon what Help says--this smells like a syntax generation bug. I'm not very knowledgeable about ARIMA and can't get you much advice. Syntax generation bugs are always easy to get around by pasting the syntax and editing it to get the model you want. -----Original Message----- From: Guerrero, Rodrigo [mailto:[hidden email]] Sent: Wednesday, April 29, 2009 3:25 PM To: ViAnn Beadle; [hidden email] Subject: RE: Re: ARIMA Modeling Thank you. I have some models that have the AR = [2,1] combination and some that have the AR = [2]. It does not seem to be consistent and I cannot figure out the difference. The models are predicting different data series, but are identical in the way they got set up in the dialog. RG Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle Sent: Wednesday, April 29, 2009 11:58 AM To: [hidden email] Subject: Re: ARIMA Modeling HELP on the TSMODEL syntax says that this specifies 1st and 2nd lags. HELP on the Time Series Modeler says the following: " All positive lower orders will be included in the model. For example, if you specify 2, the model includes orders 2 and 1." -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo Sent: Wednesday, April 29, 2009 9:48 AM To: [hidden email] Subject: ARIMA Modeling Hello all, I am doing some time series forecasting using ARIMA in V17. I used the menu to build my model, but then pasted the syntax. The model I ran had a AR value of 2, but this is what was pasted in the syntax window: /ARIMA AR=[2,1] DIFF=0 MA=[1] ARSEASONAL=[0] DIFFSEASONAL=0 MASEASONAL=[0] TRANSFORM=SQRT CONSTANT=YES Can anyone help me understand what the AR = [2,1] parameter means and how it is different from AR = [2]? Thanks. Rodrigo Guerrero Director of Marketing Research and Analysis The SCOOTER Store Office: 830-627-4317 Mobile: 830-832-8490 We provide FREEDOM and INDEPENDENCE to people with limited mobility The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. The information transmitted is intended only for the addressee(s) and may contain confidential or privileged material, or both. Any review, receipt, dissemination or other use of this information by non-addressees is prohibited. If you received this in error or are a non-addressee, please contact the sender and delete the transmitted information. ===================== 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 |
|
|
In reply to this post by Ornelas, Fermin-2
Yes, that's how it works.
Rodrigo, can you describe the steps that led to the dialog pasting AR=[2] instead of AR=[2,1]? Alex -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ornelas, Fermin Sent: Wednesday, April 29, 2009 5:09 PM To: [hidden email] Subject: Re: ARIMA Modeling I could be wrong, but it seems that [2] will run an ARIMA model with autoregressive parameter of order 2 only. The second model as ViAnn inferred should have both parameters of order 1 and 2 in the model. If you ran the model you should have more parameters in the second one. At least that is how I run mine in SAS. -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle Sent: Wednesday, April 29, 2009 3:00 PM To: [hidden email] Subject: Re: ARIMA Modeling Based upon what Help says--this smells like a syntax generation bug. I'm not very knowledgeable about ARIMA and can't get you much advice. Syntax generation bugs are always easy to get around by pasting the syntax and editing it to get the model you want. -----Original Message----- From: Guerrero, Rodrigo [mailto:[hidden email]] Sent: Wednesday, April 29, 2009 3:25 PM To: ViAnn Beadle; [hidden email] Subject: RE: Re: ARIMA Modeling Thank you. I have some models that have the AR = [2,1] combination and some that have the AR = [2]. It does not seem to be consistent and I cannot figure out the difference. The models are predicting different data series, but are identical in the way they got set up in the dialog. RG Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle Sent: Wednesday, April 29, 2009 11:58 AM To: [hidden email] Subject: Re: ARIMA Modeling HELP on the TSMODEL syntax says that this specifies 1st and 2nd lags. HELP on the Time Series Modeler says the following: " All positive lower orders will be included in the model. For example, if you specify 2, the model includes orders 2 and 1." -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo Sent: Wednesday, April 29, 2009 9:48 AM To: [hidden email] Subject: ARIMA Modeling Hello all, I am doing some time series forecasting using ARIMA in V17. I used the menu to build my model, but then pasted the syntax. The model I ran had a AR value of 2, but this is what was pasted in the syntax window: /ARIMA AR=[2,1] DIFF=0 MA=[1] ARSEASONAL=[0] DIFFSEASONAL=0 MASEASONAL=[0] TRANSFORM=SQRT CONSTANT=YES Can anyone help me understand what the AR = [2,1] parameter means and how it is different from AR = [2]? Thanks. Rodrigo Guerrero Director of Marketing Research and Analysis The SCOOTER Store Office: 830-627-4317 Mobile: 830-832-8490 ===================== 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 |
Statistics Flow-chart updated
|
|
In reply to this post by Marta Garcia-Granero
Hi everybody
Although the web page doesn't reflect it yet, the flow-chart of statistical methods has been updated (it includes now sections on agreement). I'd like to receive some feedback. Best regards, Marta García-Granero -- For miscellaneous SPSS related statistical stuff, visit: http://gjyp.nl/marta/ Updated today! ===================== 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 |
Re: Statistics Flow-chart updated
|
|
Hi Marta,
Thanks for the flowchart of statistical techniques. It is useful to have a 1-page summary like this. There are several textbooks I think that have a similar summary via a table or a decision tree (eg (from memory) Andy Field's book, Perry Hinton's SPSS Explained). One thing I wondered was whether it may be useful to include something about descriptive statistics - perhaps a rather neglected area (even if neglected, there is continuity between descriptive and inferential statistics). regards Clive. ===================== 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 |
Re: Statistics Flow-chart updated
|
|
Clive Downs wrote:
> Thanks for the flowchart of statistical techniques. It is useful to have a > 1-page summary like this. There are several textbooks I think that have a > similar summary via a table or a decision tree (eg (from memory) Andy > Field's book, Perry Hinton's SPSS Explained). > Yes, I'm quite sure I didn't invent anything new. I wrote mine from scratch (I didn't know there were several already made, perhaps I wouldn't have taken the time to prepare mine had I known that) for teaching. It was in Spanish and I translated it to English later after someone at the list asked for one. Perhaps mine has the advantages of being absolutely free (you don't have to buy any book to get it), updatable, and it can be carried (in electronic or printed format) without having to take the whole book. > One thing I wondered was whether it may be useful to include something > about descriptive statistics - perhaps a rather neglected area (even if > neglected, there is continuity between descriptive and inferential > statistics). > I always assume that the knowledge is already there: people should not try any statistical test until they know what a mean, median, standard deviation... is, the difference between an ordinal or quantitative variable... Besides, there's simply no space for descriptives in the flowchart, if I want to keep its size down to one page. Regards, Marta -- For miscellaneous SPSS related statistical stuff, visit: http://gjyp.nl/marta/ ===================== 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 |
Re: Statistics Flow-chart updated
|
|
In reply to this post by Clive Downs
Clive Downs wrote:
> One thing I wondered was whether it may be useful to include something > about descriptive statistics - perhaps a rather neglected area (even if > neglected, there is continuity between descriptive and inferential > statistics). > Just an afterthought: this is the link to a very nice document on descriptives: http://www2.sjsu.edu/faculty/gerstman/StatPrimer/sumstats.pdf No need for me to write my version, I always recommend this one to my E-MENU (European Master on Metabolism & Nutrition) students. Marta -- For miscellaneous SPSS related statistical stuff, visit: http://gjyp.nl/marta/ ===================== 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 |
Re: Statistics Flow-chart updated
|
|
In reply to this post by Marta Garcia-Granero
Hi Marta
Thanks for the link - the descriptives document looks very useful. Regards, Clive. ===================== 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 |
Re: Statistics Flow-chart updated
|
|
In reply to this post by Marta Garcia-Granero
kornbrot wrote:
> Lost original link to flow chart > Please re-supply > Best > diana Hi Diana. My signature always shows the link to the page, Marta -- For miscellaneous SPSS related statistical stuff, visit: http://gjyp.nl/marta/ <---------------- HERE IT IS ===================== 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 |
Re: Statistics Flow-chart updated
|
|
Hi Marta,
Your flowchart is great! That's for providing it.
I noticed rwg, an index of agreement commonly used in social science research, was not listed on your chart. I've included pertinent references below.
Regards,
Taylor James, L., Demaree, R. G., & Wolf, G. (1984). Estimating within group interrater reliability with and without response bias. Journal of Applied Psychology, 69, 85-98. James, L., Demaree, R. G., & Wolf, G (1993). r-sub(wg): An assessment of within-group interrater agreement. Journal of Applied Psychology, 78, 306-309. From: SPSSX(r) Discussion on behalf of Marta García-Granero Sent: Fri 5/8/2009 12:48 PM To: [hidden email] Subject: Re: Statistics Flow-chart updated kornbrot wrote: |
Re: Statistics Flow-chart updated
|
|
Poling, Taylor Leigh wrote:
> > I noticed rwg, an index of agreement commonly used in social science research, was not listed on your chart. I've included pertinent references below. > > > James, L., Demaree, R. G., & Wolf, G. (1984). Estimating within group interrater reliability with and without response bias. Journal of Applied Psychology, 69, 85-98. > > James, L., Demaree, R. G., & Wolf, G (1993). r-sub(wg): An assessment of within-group interrater agreement. Journal of Applied Psychology, 78, 306-309. > Thanks for the tip. Since I'm a biologist and work mainly with biomedical data, my experience with Rwg is null. I'll have to read a bit about it before I decide what to do with it. Marta -- For miscellaneous SPSS related statistical stuff, visit: http://gjyp.nl/marta/ ===================== 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 |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |