SPSSX Discussion

odds ratio

Classic

List

Threaded

10 messages Options

eunkum

odds ratio

I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and C19:1CE. Please see the following output. PCa is prostate cancer.

CompoundName * PCa Crosstabulation

PCa Total
Absent Present
CompoundName C18:2 CE Count 0 1 1
% within CompoundName 0.0% 100.0% 100.0%
C19:1 CE Count 1 0 1
% within CompoundName 100.0% 0.0% 100.0%
Total Count 1 1 2
% within CompoundName 50.0% 50.0% 100.0%

I got the following message that says that risk estimate statistics cannot be computed even though this is a 2*2 table without empty cells. How can I fix it?

Risk Estimate
Value
Odds Ratio for PCa (Absent / Present) a

a Risk Estimate statistics cannot be computed. They are only computed for a 2*2 table without empty cells.

Bruce Weaver

Re: odds ratio

Administrator

Your 2x2 table has two cells with observed count=0 (i.e., empty cells), as follows:

a=0 b=1
c=1 d=0

The following syntax duplicates your results, including the warning about empty cells:

DATA LIST list / CompoundName PCa (2F1).
BEGIN DATA
0 1
1 0
END DATA.

VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
CROSSTABS CompoundName BY PCa
/CELLS=COUNT ROW /STATISTICS=RISK.

If you have the 4 cell counts, try the following, replacing the letters a, b, c and d with the 4 cell counts:

DATA LIST list / CompoundName PCa (2F1) N (F5.0).
BEGIN DATA
0 0 a
0 1 b
1 0 c
1 1 d
END DATA.

VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
WEIGHT BY N.
CROSSTABS CompoundName BY PCa
/CELLS=COUNT ROW /STATISTICS=RISK.

eunkum wrote

I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and C19:1CE. Please see the following output. PCa is prostate cancer.

CompoundName * PCa Crosstabulation

PCa Total
Absent Present
CompoundName C18:2 CE Count 0 1 1
% within CompoundName 0.0% 100.0% 100.0%
C19:1 CE Count 1 0 1
% within CompoundName 100.0% 0.0% 100.0%
Total Count 1 1 2
% within CompoundName 50.0% 50.0% 100.0%

I got the following message that says that risk estimate statistics cannot be computed even though this is a 2*2 table without empty cells. How can I fix it?

Risk Estimate
Value
Odds Ratio for PCa (Absent / Present) a

a Risk Estimate statistics cannot be computed. They are only computed for a 2*2 table without empty cells.

--
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/).

eunkum

Re: odds ratio

Thanks. I ran the syntax. I got the following message.

DATASET NAME DataSet1 WINDOW=FRONT.
NEW FILE.
DATASET NAME DataSet2 WINDOW=FRONT.
DATA LIST list / CompoundName PCa (2F1) N (F5.0).
VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
WEIGHT BY N.
CROSSTABS CompoundName BY PCa

>Error # 119. Command name: CROSSTABS
>The in-line data must appear before a procedure can be executed and before the
>data can be modified in the data editor.
>Execution of this command stops.
/CELLS=COUNT ROW /STATISTICS=RISK.

On Wed, Jun 15, 2016 at 5:47 PM, Bruce Weaver [via SPSSX Discussion] <[hidden email]> wrote:

Your 2x2 table has two cells with observed count=0 (i.e., empty cells), as follows:

a=0 b=1
c=1 d=0

The following syntax duplicates your results, including the warning about empty cells:

DATA LIST list / CompoundName PCa (2F1).
BEGIN DATA
0 1
1 0
END DATA.

VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
CROSSTABS CompoundName BY PCa
/CELLS=COUNT ROW /STATISTICS=RISK.

If you have the 4 cell counts, try the following, replacing the letters a, b, c and d with the 4 cell counts:

DATA LIST list / CompoundName PCa (2F1) N (F5.0).
BEGIN DATA
0 0 a
0 1 b
1 0 c
1 1 d
END DATA.

VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
WEIGHT BY N.
CROSSTABS CompoundName BY PCa
/CELLS=COUNT ROW /STATISTICS=RISK.

eunkum wrote

I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and C19:1CE. Please see the following output. PCa is prostate cancer.

CompoundName * PCa Crosstabulation

PCa Total
Absent Present
CompoundName C18:2 CE Count 0 1 1
% within CompoundName 0.0% 100.0% 100.0%
C19:1 CE Count 1 0 1
% within CompoundName 100.0% 0.0% 100.0%
Total Count 1 1 2
% within CompoundName 50.0% 50.0% 100.0%

I got the following message that says that risk estimate statistics cannot be computed even though this is a 2*2 table without empty cells. How can I fix it?

Risk Estimate
Value
Odds Ratio for PCa (Absent / Present) a

a Risk Estimate statistics cannot be computed. They are only computed for a 2*2 table without empty cells.

--
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.

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732436.html

To unsubscribe from odds ratio, click here.
NAML

eunkum

Re: odds ratio

In reply to this post by Bruce Weaver

Thank you for your help. Is there any way for me to calculate odds ration for 31*2 table (lipids (31,e.g., C18.2CE, C19.1CE)* PCa (2, absent/present)? It's 31*2 table. Risk estimate statistics are only computed for a 2*2 table.

On Thu, Jun 16, 2016 at 11:55 AM, Eun Um <[hidden email]> wrote:

Thanks. I ran the syntax. I got the following message.

DATASET NAME DataSet1 WINDOW=FRONT.
NEW FILE.
DATASET NAME DataSet2 WINDOW=FRONT.
DATA LIST list / CompoundName PCa (2F1) N (F5.0).
VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
WEIGHT BY N.
CROSSTABS CompoundName BY PCa

>Error # 119. Command name: CROSSTABS
>The in-line data must appear before a procedure can be executed and before the
>data can be modified in the data editor.
>Execution of this command stops.
/CELLS=COUNT ROW /STATISTICS=RISK.

On Wed, Jun 15, 2016 at 5:47 PM, Bruce Weaver [via SPSSX Discussion] <[hidden email]> wrote:
Your 2x2 table has two cells with observed count=0 (i.e., empty cells), as follows:

a=0 b=1
c=1 d=0

The following syntax duplicates your results, including the warning about empty cells:

DATA LIST list / CompoundName PCa (2F1).
BEGIN DATA
0 1
1 0
END DATA.

VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
CROSSTABS CompoundName BY PCa
/CELLS=COUNT ROW /STATISTICS=RISK.

If you have the 4 cell counts, try the following, replacing the letters a, b, c and d with the 4 cell counts:

DATA LIST list / CompoundName PCa (2F1) N (F5.0).
BEGIN DATA
0 0 a
0 1 b
1 0 c
1 1 d
END DATA.

VALUE LABELS
CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
PCa 0 "Absent" 1 "Present"
.
WEIGHT BY N.
CROSSTABS CompoundName BY PCa
/CELLS=COUNT ROW /STATISTICS=RISK.

eunkum wrote

I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and C19:1CE. Please see the following output. PCa is prostate cancer.

CompoundName * PCa Crosstabulation

PCa Total
Absent Present
CompoundName C18:2 CE Count 0 1 1
% within CompoundName 0.0% 100.0% 100.0%
C19:1 CE Count 1 0 1
% within CompoundName 100.0% 0.0% 100.0%
Total Count 1 1 2
% within CompoundName 50.0% 50.0% 100.0%

I got the following message that says that risk estimate statistics cannot be computed even though this is a 2*2 table without empty cells. How can I fix it?

Risk Estimate
Value
Odds Ratio for PCa (Absent / Present) a

a Risk Estimate statistics cannot be computed. They are only computed for a 2*2 table without empty cells.

--
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.

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732436.html

To unsubscribe from odds ratio, click here.
NAML

David Marso

Re: odds ratio

Administrator

You really need to get in the habit of defining your questions with more detail and precision!!!
I don't think I need to elaborate here.

eunkum wrote

Thank you for your help. Is there any way for me to calculate odds ration
for 31*2 table (lipids (31,e.g., C18.2CE, C19.1CE)* PCa (2,
absent/present)? It's 31*2 table. Risk estimate statistics are only
computed for a 2*2 table.

On Thu, Jun 16, 2016 at 11:55 AM, Eun Um <[hidden email]> wrote:

> Thanks. I ran the syntax. I got the following message.
>
> DATASET NAME DataSet1 WINDOW=FRONT.
> NEW FILE.
> DATASET NAME DataSet2 WINDOW=FRONT.
> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
> VALUE LABELS
> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
> PCa 0 "Absent" 1 "Present"
> .
> WEIGHT BY N.
> CROSSTABS CompoundName BY PCa
>
> >Error # 119. Command name: CROSSTABS
> >The in-line data must appear before a procedure can be executed and
> before the
> >data can be modified in the data editor.
> >Execution of this command stops.
> /CELLS=COUNT ROW /STATISTICS=RISK.
>
>
> On Wed, Jun 15, 2016 at 5:47 PM, Bruce Weaver [via SPSSX Discussion] <
> [hidden email]> wrote:
>
>> Your 2x2 table has two cells with observed count=0 (i.e., empty cells),
>> as follows:
>>
>> a=0 b=1
>> c=1 d=0
>>
>> The following syntax duplicates your results, including the warning about
>> empty cells:
>>
>> DATA LIST list / CompoundName PCa (2F1).
>> BEGIN DATA
>> 0 1
>> 1 0
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>> If you have the 4 cell counts, try the following, replacing the letters
>> a, b, c and d with the 4 cell counts:
>>
>> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
>> BEGIN DATA
>> 0 0 a
>> 0 1 b
>> 1 0 c
>> 1 1 d
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> WEIGHT BY N.
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>>
>>
>> eunkum wrote
>> I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and
>> C19:1CE. Please see the following output. PCa is prostate cancer.
>>
>>
>>
>> CompoundName * PCa Crosstabulation
>>
>> PCa
>> Total
>>
>> Absent Present
>> CompoundName C18:2 CE Count 0 1 1
>> % within CompoundName 0.0%
>> 100.0% 100.0%
>> C19:1 CE Count 1 0 1
>> % within CompoundName 100.0% 0.0%
>> 100.0%
>> Total Count 1 1 2
>> % within CompoundName 50.0% 50.0%
>> 100.0%
>>
>>
>> I got the following message that says that risk estimate statistics
>> cannot be computed even though this is a 2*2 table without empty cells. How
>> can I fix it?
>>
>> Risk Estimate
>> Value
>> Odds Ratio for PCa (Absent / Present) a
>>
>> a Risk Estimate statistics cannot be computed. They are only computed for
>> a 2*2 table without empty cells.
>>
>> --
>> 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. *
>>
>>
>> ------------------------------
>> If you reply to this email, your message will be added to the discussion
>> below:
>>
>> http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732436.html
>> To unsubscribe from odds ratio, click here
>> <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=5732425&code=ZXVua3VtQGdtYWlsLmNvbXw1NzMyNDI1fC0xMjIwMzkyMzE0>
>> .
>> NAML
>> <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>>
>
>

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?"

eunkum

Re: odds ratio

I want to calculate odds ratio of 31 lipids with PCa (prostate cancer).
I want to determine odds ratio and relative risk for the 31 lipid species (Please see below). It is 31*2 table. But risk estimate statistics are only
computed for a 2*2 table. Is there any way for me to calculate risk estimate statistics for the 31 lipids species as follows?

p (lipids)      PCa
C19:1 CE    Absent
C20:0 CE    Absent
C20:1 CE    Absent
C20:2 CE    Absent
DSM(16:0    Absent
LPE(16:0    Absent
PC(38:0)    Absent
PC(40:2)    Absent
PC(40:3)    Absent
PC(40:7)    Absent
PC(42:10    Absent
PC(42:2)    Absent
PC(42:3)    Absent
PC(42:5)    Absent
PC(42:4)    Absent
PC(42:9)    Absent
PC(42:8)    Absent
ePC(36:1    Absent
ePC(36:5    Absent
ePC(38:1    Absent
ePC(38:2    Absent
ePC(38:3    Absent
ePC(38:5    Absent
ePC(38:6    Absent
ePC(40:2    Absent
ePC(40:5    Absent
ePC(40:4    Absent
ePC(40:3    Absent
ePE(34:1    Absent
ePE(36:3    Absent
C18:2 CE    Present

On Thu, Jun 16, 2016 at 12:51 PM, David Marso [via SPSSX Discussion] <[hidden email]> wrote:

You really need to get in the habit of defining your questions with more detail and precision!!!
I don't think I need to elaborate here.

eunkum wrote

Thank you for your help. Is there any way for me to calculate odds ration
for 31*2 table (lipids (31,e.g., C18.2CE, C19.1CE)* PCa (2,
absent/present)? It's 31*2 table. Risk estimate statistics are only
computed for a 2*2 table.

On Thu, Jun 16, 2016 at 11:55 AM, Eun Um <[hidden email]> wrote:

> Thanks. I ran the syntax. I got the following message.
>
> DATASET NAME DataSet1 WINDOW=FRONT.
> NEW FILE.
> DATASET NAME DataSet2 WINDOW=FRONT.
> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
> VALUE LABELS
> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
> PCa 0 "Absent" 1 "Present"
> .
> WEIGHT BY N.
> CROSSTABS CompoundName BY PCa
>
> >Error # 119. Command name: CROSSTABS
> >The in-line data must appear before a procedure can be executed and
> before the
> >data can be modified in the data editor.
> >Execution of this command stops.
> /CELLS=COUNT ROW /STATISTICS=RISK.
>
>
> On Wed, Jun 15, 2016 at 5:47 PM, Bruce Weaver [via SPSSX Discussion] <
> [hidden email]> wrote:

>
>> Your 2x2 table has two cells with observed count=0 (i.e., empty cells),
>> as follows:
>>
>> a=0 b=1
>> c=1 d=0
>>
>> The following syntax duplicates your results, including the warning about
>> empty cells:
>>
>> DATA LIST list / CompoundName PCa (2F1).
>> BEGIN DATA
>> 0 1
>> 1 0
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>> If you have the 4 cell counts, try the following, replacing the letters
>> a, b, c and d with the 4 cell counts:
>>
>> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
>> BEGIN DATA
>> 0 0 a
>> 0 1 b
>> 1 0 c
>> 1 1 d
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> WEIGHT BY N.
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>>
>>
>> eunkum wrote
>> I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and
>> C19:1CE. Please see the following output. PCa is prostate cancer.
>>
>>
>>
>> CompoundName * PCa Crosstabulation
>>
>> PCa
>> Total
>>
>> Absent Present
>> CompoundName C18:2 CE Count 0 1 1
>> % within CompoundName 0.0%
>> 100.0% 100.0%
>> C19:1 CE Count 1 0 1
>> % within CompoundName 100.0% 0.0%
>> 100.0%
>> Total Count 1 1 2
>> % within CompoundName 50.0% 50.0%
>> 100.0%
>>
>>
>> I got the following message that says that risk estimate statistics
>> cannot be computed even though this is a 2*2 table without empty cells. How
>> can I fix it?
>>
>> Risk Estimate
>> Value
>> Odds Ratio for PCa (Absent / Present) a
>>
>> a Risk Estimate statistics cannot be computed. They are only computed for
>> a 2*2 table without empty cells.
>>
>> --
>> 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. *
>>
>>
>> ------------------------------
>> If you reply to this email, your message will be added to the discussion
>> below:
>>
>> http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732436.html
>> To unsubscribe from odds ratio, click here
>> < .
>> NAML
>> <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>>
>
>

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?"

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732461.html

To unsubscribe from odds ratio, click here.
NAML

David Marso

Re: odds ratio

Administrator

Well you don't really describe the data. That would help.

Presumably each case/record has an indication of whether a given lipid was present/absent and prostate was present/absent.

We don't really know how the data are arranged or the unit of observation.

You might consider seeking out someone proficient in doing data analysis with this sort of thing before you end up with a big mess.

You ever seen the movie Cool Hand Luke?

On Thu, Jun 16, 2016 at 1:01 PM, eunkum [via SPSSX Discussion] <[hidden email]> wrote:

I want to calculate odds ratio of 31 lipids with PCa (prostate cancer).
I want to determine odds ratio and relative risk for the 31 lipid species (Please see below). It is 31*2 table. But risk estimate statistics are only
computed for a 2*2 table. Is there any way for me to calculate risk estimate statistics for the 31 lipids species as follows?

p (lipids)      PCa
C19:1 CE    Absent
C20:0 CE    Absent
C20:1 CE    Absent
C20:2 CE    Absent
DSM(16:0    Absent
LPE(16:0    Absent
PC(38:0)    Absent
PC(40:2)    Absent
PC(40:3)    Absent
PC(40:7)    Absent
PC(42:10    Absent
PC(42:2)    Absent
PC(42:3)    Absent
PC(42:5)    Absent
PC(42:4)    Absent
PC(42:9)    Absent
PC(42:8)    Absent
ePC(36:1    Absent
ePC(36:5    Absent
ePC(38:1    Absent
ePC(38:2    Absent
ePC(38:3    Absent
ePC(38:5    Absent
ePC(38:6    Absent
ePC(40:2    Absent
ePC(40:5    Absent
ePC(40:4    Absent
ePC(40:3    Absent
ePE(34:1    Absent
ePE(36:3    Absent
C18:2 CE    Present

On Thu, Jun 16, 2016 at 12:51 PM, David Marso [via SPSSX Discussion] <[hidden email]> wrote:

You really need to get in the habit of defining your questions with more detail and precision!!!
I don't think I need to elaborate here.

eunkum wrote

Thank you for your help. Is there any way for me to calculate odds ration
for 31*2 table (lipids (31,e.g., C18.2CE, C19.1CE)* PCa (2,
absent/present)? It's 31*2 table. Risk estimate statistics are only
computed for a 2*2 table.

On Thu, Jun 16, 2016 at 11:55 AM, Eun Um <[hidden email]> wrote:

> Thanks. I ran the syntax. I got the following message.
>
> DATASET NAME DataSet1 WINDOW=FRONT.
> NEW FILE.
> DATASET NAME DataSet2 WINDOW=FRONT.
> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
> VALUE LABELS
> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
> PCa 0 "Absent" 1 "Present"
> .
> WEIGHT BY N.
> CROSSTABS CompoundName BY PCa
>
> >Error # 119. Command name: CROSSTABS
> >The in-line data must appear before a procedure can be executed and
> before the
> >data can be modified in the data editor.
> >Execution of this command stops.
> /CELLS=COUNT ROW /STATISTICS=RISK.
>
>
> On Wed, Jun 15, 2016 at 5:47 PM, Bruce Weaver [via SPSSX Discussion] <
> [hidden email]> wrote:

>
>> Your 2x2 table has two cells with observed count=0 (i.e., empty cells),
>> as follows:
>>
>> a=0 b=1
>> c=1 d=0
>>
>> The following syntax duplicates your results, including the warning about
>> empty cells:
>>
>> DATA LIST list / CompoundName PCa (2F1).
>> BEGIN DATA
>> 0 1
>> 1 0
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>> If you have the 4 cell counts, try the following, replacing the letters
>> a, b, c and d with the 4 cell counts:
>>
>> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
>> BEGIN DATA
>> 0 0 a
>> 0 1 b
>> 1 0 c
>> 1 1 d
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> WEIGHT BY N.
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>>
>>
>> eunkum wrote
>> I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and
>> C19:1CE. Please see the following output. PCa is prostate cancer.
>>
>>
>>
>> CompoundName * PCa Crosstabulation
>>
>> PCa
>> Total
>>
>> Absent Present
>> CompoundName C18:2 CE Count 0 1 1
>> % within CompoundName 0.0%
>> 100.0% 100.0%
>> C19:1 CE Count 1 0 1
>> % within CompoundName 100.0% 0.0%
>> 100.0%
>> Total Count 1 1 2
>> % within CompoundName 50.0% 50.0%
>> 100.0%
>>
>>
>> I got the following message that says that risk estimate statistics
>> cannot be computed even though this is a 2*2 table without empty cells. How
>> can I fix it?
>>
>> Risk Estimate
>> Value
>> Odds Ratio for PCa (Absent / Present) a
>>
>> a Risk Estimate statistics cannot be computed. They are only computed for
>> a 2*2 table without empty cells.
>>
>> --
>> 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. *
>>
>>
>> ------------------------------
>> If you reply to this email, your message will be added to the discussion
>> below:
>>
>> http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732436.html
>> To unsubscribe from odds ratio, click here
>> < .
>> NAML
>> <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>>
>
>

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?"

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732461.html

To unsubscribe from odds ratio, click here.
NAML

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732463.html

To unsubscribe from odds ratio, click here.
NAML

eunkum

Re: odds ratio

p (lipids) consists of 31 lipids (C19:1 CE, C20:0CE, C20:1 CE, DSM(16:0)...,Please see below). PCa consists of absent/present. I have not seen the movie.

p                    PCa
C19:1 CE    Absent
C20:0 CE    Absent
C20:1 CE    Absent
C20:2 CE    Absent
DSM(16:0    Absent
LPE(16:0    Absent
PC(38:0)    Absent
PC(40:2)    Absent
PC(40:3)    Absent
PC(40:7)    Absent
PC(42:10    Absent
PC(42:2)    Absent
PC(42:3)    Absent
PC(42:5)    Absent
PC(42:4)    Absent
PC(42:9)    Absent
PC(42:8)    Absent
ePC(36:1    Absent
ePC(36:5    Absent
ePC(38:1    Absent
ePC(38:2    Absent
ePC(38:3    Absent
ePC(38:5    Absent
ePC(38:6    Absent
ePC(40:2    Absent
ePC(40:5    Absent
ePC(40:4    Absent
ePC(40:3    Absent
ePE(34:1    Absent
ePE(36:3    Absent
C18:2 CE    Present

On Thu, Jun 16, 2016 at 1:22 PM, David Marso [via SPSSX Discussion] <[hidden email]> wrote:

Well you don't really describe the data. That would help.
Presumably each case/record has an indication of whether a given lipid was present/absent and prostate was present/absent.
We don't really know how the data are arranged or the unit of observation.
You might consider seeking out someone proficient in doing data analysis with this sort of thing before you end up with a big mess.
You ever seen the movie Cool Hand Luke?

On Thu, Jun 16, 2016 at 1:01 PM, eunkum [via SPSSX Discussion] <[hidden email]> wrote:

I want to calculate odds ratio of 31 lipids with PCa (prostate cancer).
I want to determine odds ratio and relative risk for the 31 lipid species (Please see below). It is 31*2 table. But risk estimate statistics are only
computed for a 2*2 table. Is there any way for me to calculate risk estimate statistics for the 31 lipids species as follows?

p (lipids)      PCa
C19:1 CE    Absent
C20:0 CE    Absent
C20:1 CE    Absent
C20:2 CE    Absent
DSM(16:0    Absent
LPE(16:0    Absent
PC(38:0)    Absent
PC(40:2)    Absent
PC(40:3)    Absent
PC(40:7)    Absent
PC(42:10    Absent
PC(42:2)    Absent
PC(42:3)    Absent
PC(42:5)    Absent
PC(42:4)    Absent
PC(42:9)    Absent
PC(42:8)    Absent
ePC(36:1    Absent
ePC(36:5    Absent
ePC(38:1    Absent
ePC(38:2    Absent
ePC(38:3    Absent
ePC(38:5    Absent
ePC(38:6    Absent
ePC(40:2    Absent
ePC(40:5    Absent
ePC(40:4    Absent
ePC(40:3    Absent
ePE(34:1    Absent
ePE(36:3    Absent
C18:2 CE    Present

On Thu, Jun 16, 2016 at 12:51 PM, David Marso [via SPSSX Discussion] <[hidden email]> wrote:

You really need to get in the habit of defining your questions with more detail and precision!!!
I don't think I need to elaborate here.

eunkum wrote

Thank you for your help. Is there any way for me to calculate odds ration
for 31*2 table (lipids (31,e.g., C18.2CE, C19.1CE)* PCa (2,
absent/present)? It's 31*2 table. Risk estimate statistics are only
computed for a 2*2 table.

On Thu, Jun 16, 2016 at 11:55 AM, Eun Um <[hidden email]> wrote:

> Thanks. I ran the syntax. I got the following message.
>
> DATASET NAME DataSet1 WINDOW=FRONT.
> NEW FILE.
> DATASET NAME DataSet2 WINDOW=FRONT.
> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
> VALUE LABELS
> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
> PCa 0 "Absent" 1 "Present"
> .
> WEIGHT BY N.
> CROSSTABS CompoundName BY PCa
>
> >Error # 119. Command name: CROSSTABS
> >The in-line data must appear before a procedure can be executed and
> before the
> >data can be modified in the data editor.
> >Execution of this command stops.
> /CELLS=COUNT ROW /STATISTICS=RISK.
>
>
> On Wed, Jun 15, 2016 at 5:47 PM, Bruce Weaver [via SPSSX Discussion] <
> [hidden email]> wrote:

>
>> Your 2x2 table has two cells with observed count=0 (i.e., empty cells),
>> as follows:
>>
>> a=0 b=1
>> c=1 d=0
>>
>> The following syntax duplicates your results, including the warning about
>> empty cells:
>>
>> DATA LIST list / CompoundName PCa (2F1).
>> BEGIN DATA
>> 0 1
>> 1 0
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>> If you have the 4 cell counts, try the following, replacing the letters
>> a, b, c and d with the 4 cell counts:
>>
>> DATA LIST list / CompoundName PCa (2F1) N (F5.0).
>> BEGIN DATA
>> 0 0 a
>> 0 1 b
>> 1 0 c
>> 1 1 d
>> END DATA.
>>
>> VALUE LABELS
>> CompoundName 0 "C18:2 CE" 1 "C19:1 CE" /
>> PCa 0 "Absent" 1 "Present"
>> .
>> WEIGHT BY N.
>> CROSSTABS CompoundName BY PCa
>> /CELLS=COUNT ROW /STATISTICS=RISK.
>>
>>
>>
>>
>> eunkum wrote
>> I want to calculate odds ratio for two lipids (biomakrers) - C18:2 CE and
>> C19:1CE. Please see the following output. PCa is prostate cancer.
>>
>>
>>
>> CompoundName * PCa Crosstabulation
>>
>> PCa
>> Total
>>
>> Absent Present
>> CompoundName C18:2 CE Count 0 1 1
>> % within CompoundName 0.0%
>> 100.0% 100.0%
>> C19:1 CE Count 1 0 1
>> % within CompoundName 100.0% 0.0%
>> 100.0%
>> Total Count 1 1 2
>> % within CompoundName 50.0% 50.0%
>> 100.0%
>>
>>
>> I got the following message that says that risk estimate statistics
>> cannot be computed even though this is a 2*2 table without empty cells. How
>> can I fix it?
>>
>> Risk Estimate
>> Value
>> Odds Ratio for PCa (Absent / Present) a
>>
>> a Risk Estimate statistics cannot be computed. They are only computed for
>> a 2*2 table without empty cells.
>>
>> --
>> 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. *
>>
>>
>> ------------------------------
>> If you reply to this email, your message will be added to the discussion
>> below:
>>
>> http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732436.html
>> To unsubscribe from odds ratio, click here
>> < .
>> NAML
>> <http://spssx-discussion.1045642.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>>
>
>

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?"

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732461.html

To unsubscribe from odds ratio, click here.
NAML

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732463.html

To unsubscribe from odds ratio, click here.
NAML

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?"

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/odds-ratio-tp5732425p5732464.html

To unsubscribe from odds ratio, click here.
NAML

David Marso

Re: odds ratio

Administrator

Very briefly and then I will join Gene in being done with this thread until something which is actually useful to assist is posted.

You must find a way to create a 2 x 2 table for each Lipid (1..31) with frequencies of presence/absence of lipid(I) in row and presence/absence of PC in the columns. If you are dealing with aggregated summaries without presence/absence of the two crossed then good luck (yer SOL) -unless you have access to the raw data-.
"What we've got here is failure to communicate"
https://www.youtube.com/watch?v=452XjnaHr1A

PRogman

Re: odds ratio

In reply to this post by eunkum

If you only have 3 cases with cancer (PCa>1) in 374 cases, further, all p (lipids) are unique.
I doubt you will get very much out of this dataset, unless you have more information.

i Lipid PCa
5 C18:1 CE 2,52
6 C18:2 CE 15,38
14 C20:4 CE 3,31

Recommend this IDRE UCLA page: http://www.ats.ucla.edu/stat/mult_pkg/faq/general/odds_ratio.htm

/PRogman