Quantile Regression in SPSS 26 (could someone help?)

[Marcos Colla]

I would like to apologize for my limitation in English, so, if I say
something wrong or something that you cannot understand I ask you to please
forgive me. I'm a master’s degree student in the Master Program in
Production and Systems Engineering and I'm working hard in my master thesis.
My research project is focused on understanding the ambulances response time
(dependent variable) in the local Emergency Medical System. After sistematic
literature review I found the interesting possibility of quantile regression
(available on SPSS 26). So,  I need to understand the change in a specific
percentile of dependent variable caused by changing a independent variable
(e.g. hour of the day, day of the week and many others). I'm trying to find
a step by step to use quantile regression on SPSS 26. Somebody can help me,
please?



--
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]

There are two quantile regression procedures  available in Statistics V26.  The QUANTILE REGRESSION procedure is built in and should be used rather than the older SPSSINC QUANTREG extension command available via Extensions > Extension Hub with one exception.

To get a better understanding of quantile regression, start with the case study in the help system (Help > Table of Contents > Case Studies > Standard Edition > Quantile Regression (https://www.ibm.com/support/knowledgecenter/en/SSLVMB_26.0.0/statistics_casestudies_project_ddita/spss/tutorials/quantile_regression.html)

You might also look at

Journal of Economic Perspectives—Volume 15, Number 4—Fall 2001—Pages 143–156   Koenker and Hallock, Quantile Regression

The exception I referred to above is that the QUANTILE REGRESSION procedure does not include a way to test whether the the various quantile lines actually have the same slope and, therefore, ordinary regression will suffice.  (Starting with ordinary regression is probably a good idea, anyway).  The SPSSINC QUANTREG procedure does provide such a test.

On Sat, Feb 8, 2020 at 7:35 PM Marcos Colla <[hidden email]> wrote:

I would like to apologize for my limitation in English, so, if I say
something wrong or something that you cannot understand I ask you to please
forgive me. I'm a master’s degree student in the Master Program in
Production and Systems Engineering and I'm working hard in my master thesis.
My research project is focused on understanding the ambulances response time
(dependent variable) in the local Emergency Medical System. After sistematic
literature review I found the interesting possibility of quantile regression
(available on SPSS 26). So,  I need to understand the change in a specific
percentile of dependent variable caused by changing a independent variable
(e.g. hour of the day, day of the week and many others). I'm trying to find
a step by step to use quantile regression on SPSS 26. Somebody can help me,
please?



--
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

[Marcos Colla]

Thanks Jon,

I was looking for just that.

Marcos



--
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

[Bruce Weaver]

Hi Jon.  I was pleased to see the native SPSS command QUANTILE REGRESSION
added in v26.  I have not had a chance to use it yet, as my university is on
v25 for now.  Do you know if there are plans to add the missing
functionality you described below?  

Bruce


Jon Peck wrote
> --- snip ---
>
> The exception I referred to above is that the QUANTILE REGRESSION
> procedure
> does not include a way to test whether the the various quantile lines
> actually have the same slope and, therefore, ordinary regression will
> suffice.  (Starting with ordinary regression is probably a good idea,
> anyway).  The SPSSINC QUANTREG procedure does provide such a test.





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
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]

I asked for that at the time of the V26 beta, but it was not added in 26.  I hope it will appear in V27, but I have no information on that.

On Mon, Feb 10, 2020 at 7:29 AM Bruce Weaver <[hidden email]> wrote:

Hi Jon.  I was pleased to see the native SPSS command QUANTILE REGRESSION
added in v26.  I have not had a chance to use it yet, as my university is on
v25 for now.  Do you know if there are plans to add the missing
functionality you described below? 

Bruce


Jon Peck wrote
> --- snip ---
>
> The exception I referred to above is that the QUANTILE REGRESSION
> procedure
> does not include a way to test whether the the various quantile lines
> actually have the same slope and, therefore, ordinary regression will
> suffice.  (Starting with ordinary regression is probably a good idea,
> anyway).  The SPSSINC QUANTREG procedure does provide such a test.





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
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

[Bruce Weaver]

Apologies if this post appears more than once.  I first posted several hours
ago, and have still not seen my post appear in the Nabble archive.  (And my
posts via Nabble typically appear immediately.)  
-----------------------------------------

Hi Jon.  I was pleased to see the native SPSS command QUANTILE REGRESSION
added in v26.  I have not had a chance to use it yet, as my university is on
v25 for now.  Do you know if there are plans to add the missing
functionality you described below?  

Bruce


Jon Peck wrote
> --- snip ---
> The exception I referred to above is that the QUANTILE REGRESSION
> procedure
> does not include a way to test whether the the various quantile lines
> actually have the same slope and, therefore, ordinary regression will
> suffice.  (Starting with ordinary regression is probably a good idea,
> anyway).  The SPSSINC QUANTREG procedure does provide such a test.





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
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

[Bruce Weaver]

It appears to have been a (temporary) problem with Nabble.  All seems to be
working normally again.  

Cheers,
Bruce


Bruce Weaver wrote

> Apologies if this post appears more than once.  I first posted several
> hours
> ago, and have still not seen my post appear in the Nabble archive.  (And
> my
> posts via Nabble typically appear immediately.)  
> -----------------------------------------
>
> Hi Jon.  I was pleased to see the native SPSS command QUANTILE REGRESSION
> added in v26.  I have not had a chance to use it yet, as my university is
> on
> v25 for now.  Do you know if there are plans to add the missing
> functionality you described below?  
>
> Bruce
>
>
> Jon Peck wrote
>> --- snip ---
>> The exception I referred to above is that the QUANTILE REGRESSION
>> procedure
>> does not include a way to test whether the the various quantile lines
>> actually have the same slope and, therefore, ordinary regression will
>> suffice.  (Starting with ordinary regression is probably a good idea,
>> anyway).  The SPSSINC QUANTREG procedure does provide such a test.
>
>
>
>
>
> -----
> --
> Bruce Weaver

> bweaver@

> http://sites.google.com/a/lakeheadu.ca/bweaver/
>
> "When all else fails, RTFM."
>
> NOTE: My Hotmail account is not monitored regularly.
> To send me an e-mail, please use the address shown above.
>
> --
> 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





-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
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