Estimation Method for logistic regression in SPSS

[Hesham Abdel-Meguid Abdalla]

Dear all
I have an urgent question. I need to know the coefficients
estimation method for logistic regression used in SPSS 14 and later. the
method used in these versions solved the problem of spase data , but i did
not discover which method used. Can any one help me please mention a
refrence.


--
Hesham Abdel-Meguid
Statistics
Dept.
Faculty of Commerce
Assiut University

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

[Hector Maletta]

Look at the algorithms that are included in the software CD. You can also
download them from SPSS.com.
Hector

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Hesham Abdel-Meguid Abdalla
Sent: 11 September 2008 14:37
To: [hidden email]
Subject: Estimation Method for logistic regression in SPSS

Dear all
I have an urgent question. I need to know the coefficients
estimation method for logistic regression used in SPSS 14 and later. the
method used in these versions solved the problem of spase data , but i did
not discover which method used. Can any one help me please mention a
refrence.


--
Hesham Abdel-Meguid
Statistics
Dept.
Faculty of Commerce
Assiut University

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

[Reutter, Alex]

In v15 and later, they are also part of the main help system.  From the menus, Help > Topics; the Algorithms are at the bottom of the help table of contents.

Alex

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Hector Maletta
Sent: Thursday, September 11, 2008 12:29 PM
To: [hidden email]
Subject: Re: Estimation Method for logistic regression in SPSS

Look at the algorithms that are included in the software CD. You can also
download them from SPSS.com.
Hector

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Hesham Abdel-Meguid Abdalla
Sent: 11 September 2008 14:37
To: [hidden email]
Subject: Estimation Method for logistic regression in SPSS

Dear all
I have an urgent question. I need to know the coefficients
estimation method for logistic regression used in SPSS 14 and later. the
method used in these versions solved the problem of spase data , but i did
not discover which method used. Can any one help me please mention a
refrence.


--
Hesham Abdel-Meguid
Statistics
Dept.
Faculty of Commerce
Assiut University

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

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

[Hesham Abdel-Meguid Abdalla]

Dear sir

I took a look at SPSS Algorithms but i did not find the answer. my
question is "What is the Maximum likelihood approach is used by SPSS
for logistic regression? is it the conditional ML, bais correction, or
what?


In v15 and later, they are also part of the main
help system.
From the
menus, Help > Topics; the Algorithms
are at the bottom of the help table
of contents.

Alex


-----Original Message-----
From: SPSSX(r) Discussion
[mailto:[hidden email]] On Behalf Of
Hector Maletta
Sent: Thursday, September 11, 2008 12:29 PM
To:
[hidden email]
Subject: Re: Estimation Method for logistic
regression in SPSS

Look at the algorithms that are included in
the software CD. You can also
download them from SPSS.com.
Hector

-----Original Message-----
From: SPSSX(r)
Discussion [mailto:[hidden email]] On Behalf Of
Hesham
Abdel-Meguid Abdalla
Sent: 11 September 2008 14:37
To:
[hidden email]
Subject: Estimation Method for logistic
regression in SPSS

Dear all
I have an urgent question. I
need to know the coefficients
estimation method for logistic
regression used in SPSS 14 and later. the
method used in these
versions solved the problem of spase data , but i did
not discover
which method used. Can any one help me please mention a
refrence.



--
Hesham Abdel-Meguid
Statistics
Dept.
Faculty of Commerce
Assiut University

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

=====================
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
_ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_
This
message has been scanned by AUN Spam E-mail Filter and Virus
detection software.




--
Hesham
Abdel-Meguid
Statistics Dept.
Faculty of Commerce
Assiut
University

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

[Reutter, Alex]

I'm looking at the "Model" and "Maximum Likelihood Estimates (MLE)" sections in the Logistic Regression Algorithms, which explain that the estimated regression coefficients maximize the joint likelihood, and are obtained using a Newton-Raphson algorithm.  Is that what you're looking for, or do I misunderstand?

Alex


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Hesham Abdel-Meguid Abdalla
Sent: Thursday, September 11, 2008 5:24 PM
To: [hidden email]
Subject: Re: Estimation Method for logistic regression in SPSS

Dear sir

I took a look at SPSS Algorithms but i did not find the answer. my
question is "What is the Maximum likelihood approach is used by SPSS
for logistic regression? is it the conditional ML, bais correction, or
what?


In v15 and later, they are also part of the main
help system.
From the
menus, Help > Topics; the Algorithms
are at the bottom of the help table
of contents.

Alex


-----Original Message-----
From: SPSSX(r) Discussion
[mailto:[hidden email]] On Behalf Of
Hector Maletta
Sent: Thursday, September 11, 2008 12:29 PM
To:
[hidden email]
Subject: Re: Estimation Method for logistic
regression in SPSS

Look at the algorithms that are included in
the software CD. You can also
download them from SPSS.com.
Hector

-----Original Message-----
From: SPSSX(r)
Discussion [mailto:[hidden email]] On Behalf Of
Hesham
Abdel-Meguid Abdalla
Sent: 11 September 2008 14:37
To:
[hidden email]
Subject: Estimation Method for logistic
regression in SPSS

Dear all
I have an urgent question. I
need to know the coefficients
estimation method for logistic
regression used in SPSS 14 and later. the
method used in these
versions solved the problem of spase data , but i did
not discover
which method used. Can any one help me please mention a
refrence.



--
Hesham Abdel-Meguid
Statistics
Dept.
Faculty of Commerce
Assiut University

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

=====================
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
_ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_
This
message has been scanned by AUN Spam E-mail Filter and Virus
detection software.




--
Hesham
Abdel-Meguid
Statistics Dept.
Faculty of Commerce
Assiut
University

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

[Henrik Lolle-2]

And by using the GENLIN procedure you have the option to use
Newton-Raphson, Fisher scoring, or a hybrid method. I don't know if this
will be of any help.

Best,
Henrik

Reutter, Alex wrote:

> I'm looking at the "Model" and "Maximum Likelihood Estimates (MLE)" sections in the Logistic Regression Algorithms, which explain that the estimated regression coefficients maximize the joint likelihood, and are obtained using a Newton-Raphson algorithm.  Is that what you're looking for, or do I misunderstand?
>
> Alex
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Hesham Abdel-Meguid Abdalla
> Sent: Thursday, September 11, 2008 5:24 PM
> To: [hidden email]
> Subject: Re: Estimation Method for logistic regression in SPSS
>
> Dear sir
>
> I took a look at SPSS Algorithms but i did not find the answer. my
> question is "What is the Maximum likelihood approach is used by SPSS
> for logistic regression? is it the conditional ML, bais correction, or
> what?
>
>
> In v15 and later, they are also part of the main
> help system.
> From the
> menus, Help > Topics; the Algorithms
> are at the bottom of the help table
> of contents.
>
> Alex
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion
> [mailto:[hidden email]] On Behalf Of
> Hector Maletta
> Sent: Thursday, September 11, 2008 12:29 PM
> To:
> [hidden email]
> Subject: Re: Estimation Method for logistic
> regression in SPSS
>
> Look at the algorithms that are included in
> the software CD. You can also
> download them from SPSS.com.
> Hector
>
> -----Original Message-----
> From: SPSSX(r)
> Discussion [mailto:[hidden email]] On Behalf Of
> Hesham
> Abdel-Meguid Abdalla
> Sent: 11 September 2008 14:37
> To:
> [hidden email]
> Subject: Estimation Method for logistic
> regression in SPSS
>
> Dear all
> I have an urgent question. I
> need to know the coefficients
> estimation method for logistic
> regression used in SPSS 14 and later. the
> method used in these
> versions solved the problem of spase data , but i did
> not discover
> which method used. Can any one help me please mention a
> refrence.
>
>
>
> --
> Hesham Abdel-Meguid
> Statistics
> Dept.
> Faculty of Commerce
> Assiut University
>
> =====================
> 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
>
> =====================
> 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
> _ _ _ _ _ _ _ _ _
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
> _
> This
> message has been scanned by AUN Spam E-mail Filter and Virus
> detection software.
>
>
>
>
> --
> Hesham
> Abdel-Meguid
> Statistics Dept.
> Faculty of Commerce
> Assiut
> University
>
> =====================
> 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

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

[Henrik Lolle]

And if you use the GENLIN procedure you have the option to use
Newton-Raphson, Fisher scoring, or a hybrid method. I don't know if
this is of any help.

Best,
Henrik

Quoting "Reutter, Alex" <[hidden email]>:

> I'm looking at the "Model" and "Maximum Likelihood Estimates (MLE)"
> sections in the Logistic Regression Algorithms, which explain that
> the estimated regression coefficients maximize the joint likelihood,
> and are obtained using a Newton-Raphson algorithm.  Is that what
> you're looking for, or do I misunderstand?
>
> Alex
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf
> Of Hesham Abdel-Meguid Abdalla
> Sent: Thursday, September 11, 2008 5:24 PM
> To: [hidden email]
> Subject: Re: Estimation Method for logistic regression in SPSS
>
> Dear sir
>
> I took a look at SPSS Algorithms but i did not find the answer. my
> question is "What is the Maximum likelihood approach is used by SPSS
> for logistic regression? is it the conditional ML, bais correction, or
> what?
>
>
> In v15 and later, they are also part of the main
> help system.
> From the
> menus, Help > Topics; the Algorithms
> are at the bottom of the help table
> of contents.
>
> Alex
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion
> [mailto:[hidden email]] On Behalf Of
> Hector Maletta
> Sent: Thursday, September 11, 2008 12:29 PM
> To:
> [hidden email]
> Subject: Re: Estimation Method for logistic
> regression in SPSS
>
> Look at the algorithms that are included in
> the software CD. You can also
> download them from SPSS.com.
> Hector
>
> -----Original Message-----
> From: SPSSX(r)
> Discussion [mailto:[hidden email]] On Behalf Of
> Hesham
> Abdel-Meguid Abdalla
> Sent: 11 September 2008 14:37
> To:
> [hidden email]
> Subject: Estimation Method for logistic
> regression in SPSS
>
> Dear all
> I have an urgent question. I
> need to know the coefficients
> estimation method for logistic
> regression used in SPSS 14 and later. the
> method used in these
> versions solved the problem of spase data , but i did
> not discover
> which method used. Can any one help me please mention a
> refrence.
>
>
>
> --
> Hesham Abdel-Meguid
> Statistics
> Dept.
> Faculty of Commerce
> Assiut University
>
> =====================
> 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
>
> =====================
> 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
> _ _ _ _ _ _ _ _ _
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
> _
> This
> message has been scanned by AUN Spam E-mail Filter and Virus
> detection software.
>
>
>
>
> --
> Hesham
> Abdel-Meguid
> Statistics Dept.
> Faculty of Commerce
> Assiut
> University
>
> =====================
> 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
>

************************************************************
Henrik Lolle
Department of Economics, Politics and Public Administration
Aalborg University
Fibigerstraede 1
9200 Aalborg
Phone: [phone]
************************************************************

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

[Hesham Abdel-Meguid Abdalla]

I appreciate your kind help. but  i am sorry that i did not explian
well my question. MLE was developed to solve the problems of small sample
and sparse data. there is Exact appproach, Conditional ML , and penalized
ML. i need to know which method is used by SPSS to compare its
results with another one.

And if you use the GENLIN procedure
you have the option to use
Newton-Raphson, Fisher scoring, or a
hybrid method. I don't know if
this is of any help.

Best,
Henrik

Quoting "Reutter, Alex"
<[hidden email]>:

> I'm looking at the
"Model" and "Maximum Likelihood Estimates (MLE)"
> sections in the Logistic Regression Algorithms, which explain that

> the estimated regression coefficients maximize the joint
likelihood,
> and are obtained using a Newton-Raphson algorithm.
Is that what
> you're looking for, or do I misunderstand?
>
> Alex
>
>
> -----Original
Message-----
>
From: SPSSX(r) Discussion
[mailto:[hidden email]] On Behalf
> Of Hesham
Abdel-Meguid Abdalla
> Sent: Thursday, September 11, 2008 5:24 PM

> To: [hidden email]
> Subject: Re: Estimation
Method for logistic regression in SPSS
>
> Dear sir
>
> I took a look at SPSS Algorithms but i did not find the
answer. my
> question is "What is the Maximum likelihood
approach is used by SPSS
> for logistic regression? is it the
conditional ML, bais correction, or
> what?
>
>

> In v15 and later, they are also part of the main
>
help system.
>
From the
> menus, Help > Topics;
the Algorithms
> are at the bottom of the help table
>
of contents.
>
> Alex
>
>
>
-----Original Message-----
>
From: SPSSX(r) Discussion
> [mailto:[hidden email]] On Behalf Of
> Hector
Maletta
> Sent: Thursday, September 11, 2008 12:29 PM
>
To:
> [hidden email]
> Subject: Re: Estimation
Method for logistic
> regression in SPSS
>
>
Look at the algorithms that are included in
> the software CD.
You can also
> download them from SPSS.com.
> Hector
>
> -----Original Message-----
>
From:
SPSSX(r)
> Discussion [mailto:[hidden email]] On Behalf
Of
> Hesham
> Abdel-Meguid Abdalla
> Sent: 11
September 2008 14:37
> To:
> [hidden email]

> Subject: Estimation Method for logistic
> regression
in SPSS
>
> Dear all
> I have an urgent
question. I
> need to know the coefficients
> estimation
method for logistic
> regression used in SPSS 14 and later. the

> method used in these
> versions solved the problem of
spase data , but i did
> not discover
> which method
used. Can any one help me please mention a
> refrence.
>

>
>
> --
> Hesham Abdel-Meguid
> Statistics
> Dept.
> Faculty of Commerce
> Assiut University
>
> =====================
> 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

>
> =====================
> 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

> _ _ _ _ _ _ _ _ _
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _
> _
> This
> message has
been scanned by AUN Spam E-mail Filter and Virus
> detection
software.

>
>
>
>
> --
> Hesham
> Abdel-Meguid
> Statistics Dept.
> Faculty of Commerce
> Assiut
> University
>
> =====================
> 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
>



************************************************************
Henrik Lolle
Department of Economics, Politics and Public
Administration
Aalborg University
Fibigerstraede 1
9200
Aalborg
Phone: [phone]
************************************************************

=====================
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
_ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_
This
message has been scanned by AUN Spam E-mail Filter and Virus
detection software.




--
Hesham
Abdel-Meguid
Statistics Dept.
Faculty of Commerce
Assiut
University

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

[SR Millis-3]

If you want an "exact" method, it is available from Cytel as LogXact.  It is also available in SAS.  There are many different penalized methods---some are available in R.  Check out Frank Harrell's book, "Regression Modeling Strategies," for details.  Stata also features some penalized methods.


Scott R Millis, PhD, MEd, ABPP (CN,CL,RP), CStat
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email:  [hidden email]
Tel: [phone]
Fax: [phone]


--- On Fri, 9/12/08, Hesham Abdel-Meguid Abdalla <[hidden email]> wrote:

> From: Hesham Abdel-Meguid Abdalla <[hidden email]>
> Subject: Re: Estimation Method for logistic regression in SPSS
> To: [hidden email]
> Date: Friday, September 12, 2008, 7:18 PM
> I appreciate your kind help. but  i am sorry that i did not
> explian
> well my question. MLE was developed to solve the problems
> of small sample
> and sparse data. there is Exact appproach, Conditional ML ,
> and penalized
> ML. i need to know which method is used by SPSS to compare
> its
> results with another one.
>

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

[Reutter, Alex]

It's just using plain old (unconditional) ML.  In the case of sparse data, the algorithm may actually not converge due to quasi-complete separation of the data - check the warnings and the iteration history table.

 

Alex

 

________________________________

From: Hesham Abdel-Meguid Abdalla [mailto:[hidden email]]
Sent: Friday, September 12, 2008 6:16 PM
To: Reutter, Alex
Subject: [BULK] Re: Estimation Method for logistic regression in SPSS
Importance: Low

 

Dear sir
I appreciate your kind help. but  i am sorry that i did not explian well my question. MLE was developed to solve the problems of small sample and sparse data. there is Exact appproach, Conditional ML , and penalized ML. i need to know which method is used by SPSS to compare its results with another one.

I'm looking at the "Model" and "Maximum Likelihood Estimates (MLE)"
sections in the Logistic Regression Algorithms, which explain that the
estimated regression coefficients maximize the joint likelihood, and are
obtained using a Newton-Raphson algorithm. Is that what you're looking
for, or do I misunderstand?

Alex


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Hesham Abdel-Meguid Abdalla
Sent: Thursday, September 11, 2008 5:24 PM
To: [hidden email]
Subject: Re: Estimation Method for logistic regression in SPSS

Dear sir

I took a look at SPSS Algorithms but i did not find the answer. my
question is "What is the Maximum likelihood approach is used by SPSS
for logistic regression? is it the conditional ML, bais correction, or
what?


In v15 and later, they are also part of the main
help system.
>From the
menus, Help > Topics; the Algorithms
are at the bottom of the help table
of contents.

Alex


-----Original Message-----
From: SPSSX(r) Discussion
[mailto:[hidden email]] On Behalf Of
Hector Maletta
Sent: Thursday, September 11, 2008 12:29 PM
To:
[hidden email]
Subject: Re: Estimation Method for logistic
regression in SPSS

Look at the algorithms that are included in
the software CD. You can also
download them from SPSS.com.
Hector

-----Original Message-----
From: SPSSX(r)
Discussion [mailto:[hidden email]] On Behalf Of
Hesham
Abdel-Meguid Abdalla
Sent: 11 September 2008 14:37
To:
[hidden email]
Subject: Estimation Method for logistic
regression in SPSS

Dear all
I have an urgent question. I
need to know the coefficients
estimation method for logistic
regression used in SPSS 14 and later. the
method used in these
versions solved the problem of spase data , but i did
not discover
which method used. Can any one help me please mention a
refrence.

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