SPSSX Discussion

SPSS 16.0 lag

Classic

List

Threaded

12 messages Options

Chris Traynor

SPSS 16.0 lag

IT installed SPSS v16 the other day...why does it lag when I scroll across
my variables now?

Chris

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

Eric Zylstra

Re: SPSS 16.0 lag

Java.

16 has had its interface rewritten in Java. Anything written in Java is much slower than programming code written specifically for a given platform. For me, machines that ran 14 without effort now cycle their fans to high speed under the stress of running Java.

The rewrite was a huge coding job, and I'd be surpised if SPSS was able to do any optimizing. It is hard to program well, especially on a tight release cycle. Add to that writing everything new from scratch. "Admirable" is how I'd word the accomplishment of turning out a functioning yet entirely new interface in the normal release cycle following 15.

This switch will enable SPSS to release one version for all platforms. That efficiency will then be turned into making an even better SPSS for everyone.

Eric

>>> On 5/16/2008 at 9:21 AM, in message
<[hidden email]>, Chris Traynor
<[hidden email]> wrote:

> IT installed SPSS v16 the other day...why does it lag when I scroll across
> my variables now?
>
> Chris
>
> =====================
> 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

Cleber Carmo

SPSS 16.0 - GAM models

Dear all,

Does anyone know if SPSS 16 does Generalized Additive Models analyses (poisson regression)??

Best wishes,

Cleber.

Eric Zylstra <[hidden email]> escreveu:
Java.

16 has had its interface rewritten in Java. Anything written in Java is much slower than programming code written specifically for a given platform. For me, machines that ran 14 without effort now cycle their fans to high speed under the stress of running Java.

The rewrite was a huge coding job, and I'd be surpised if SPSS was able to do any optimizing. It is hard to program well, especially on a tight release cycle. Add to that writing everything new from scratch. "Admirable" is how I'd word the accomplishment of turning out a functioning yet entirely new interface in the normal release cycle following 15.

This switch will enable SPSS to release one version for all platforms. That efficiency will then be turned into making an even better SPSS for everyone.

Eric

>>> On 5/16/2008 at 9:21 AM, in message
<[hidden email]>, Chris Traynor
wrote:

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

---------------------------------
Abra sua conta no Yahoo! Mail, o único sem limite de espaço para armazenamento!

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

Art Kendall

Re: SPSS 16.0 - GAM models

It usually better to create a new post to the list rather than
introducing a new topic via "reply".

SPSS has poisson regression. See the documentation and tutorial that
comes with the software.

Art Kendall
Social Research Consultants

Cleber Carmo wrote:

> Dear all,
>
> Does anyone know if SPSS 16 does Generalized Additive Models analyses (poisson regression)??
>
> Best wishes,
>
> Cleber.
>
> Eric Zylstra <[hidden email]> escreveu:
> Java.
>
> 16 has had its interface rewritten in Java. Anything written in Java is much slower than programming code written specifically for a given platform. For me, machines that ran 14 without effort now cycle their fans to high speed under the stress of running Java.
>
> The rewrite was a huge coding job, and I'd be surpised if SPSS was able to do any optimizing. It is hard to program well, especially on a tight release cycle. Add to that writing everything new from scratch. "Admirable" is how I'd word the accomplishment of turning out a functioning yet entirely new interface in the normal release cycle following 15.
>
> This switch will enable SPSS to release one version for all platforms. That efficiency will then be turned into making an even better SPSS for everyone.
>
> Eric
>
>
>
>>>> On 5/16/2008 at 9:21 AM, in message
>>>>
> <[hidden email]>, Chris Traynor
> wrote:
>
>> IT installed SPSS v16 the other day...why does it lag when I scroll across
>> my variables now?
>>
>> Chris
>>
>> =====================
>> 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
>
>
>
>
>
> ---------------------------------
> Abra sua conta no Yahoo! Mail, o único sem limite de espaço para armazenamento!
>
> =====================
> 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
>
>
>

Art Kendall
Social Research Consultants

Eric Janssen

Re: SPSS 16.0 - GAM models

> > Does anyone know if SPSS 16 does Generalized Additive Models analyses
> (poisson regression)??

Just a detail: Poisson regression belongs to the GLM family, not the GAM's.

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

Daniel Robertson

Re: SPSS 16.0 - GAM models

In reply to this post by Cleber Carmo

SPSS doesn't *natively* do GAM of any kind (yet -- we can hope!). GENLIN
can handle poisson distributions, but doesn't do GAM. But if you have
v.16 and the R plugin, you could probably use something like the gam()
procedure in R.

Cleber Carmo wrote:
> Dear all,
>
> Does anyone know if SPSS 16 does Generalized Additive Models analyses (poisson regression)??
>
> Best wishes,
>
> Cleber.
>
>

--
Daniel Robertson
Senior Research and Planning Associate
Institutional Research and Planning
Cornell University / irp.cornell.edu

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

Daniel Robertson

Re: SPSS 16.0 - GAM models

In reply to this post by Eric Janssen

... although more "generally" it's perfectly possible to have a GAM with
a poisson distribution -- the "G" in GAM is for generalized, after all :)

Eric Janssen wrote:
>
> Just a detail: Poisson regression belongs to the GLM family, not the GAM's.
>
>
>

--
Daniel Robertson
Senior Research and Planning Associate
Institutional Research and Planning
Cornell University / irp.cornell.edu

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

Peck, Jon

Re: SPSS 16.0 - GAM models

In reply to this post by Cleber Carmo

From the GEE Help.

Model Types
Scale Response.

• Linear. Specifies Normal as the distribution and Identity as the link function.

• Gamma with log link. Specifies Gamma as the distribution and Log as the link function.

Ordinal Response.

• Ordinal logistic. Specifies Multinomial (ordinal) as the distribution and Cumulative logit as the link function.

• Ordinal probit. Specifies Multinomial (ordinal) as the distribution and Cumulative probit as the link function.

Counts.

• Poisson loglinear. Specifies Poisson as the distribution and Log as the link function.

• Negative binomial with log link. Specifies Negative binomial (with a value of 1 for the ancillary parameter) as the distribution and Log as the link function.

Binary Response or Events/Trials Data.

• Binary logistic. Specifies Binomial as the distribution and Logit as the link function.

• Binary probit. Specifies Binomial as the distribution and Probit as the link function.

• Interval censored survival. Specifies Binomial as the distribution and Complementary log-log as the link function.

Mixture.

• Tweedie with log link. Specifies Tweedie as the distribution and Log as the link function.

• Tweedie with identity link. Specifies Tweedie as the distribution and Identity as the link function.

Custom. Specify your own combination of distribution and link function.

HTH,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Cleber Carmo
Sent: Friday, May 16, 2008 9:18 AM
To: [hidden email]
Subject: [SPSSX-L] SPSS 16.0 - GAM models

Dear all,

Does anyone know if SPSS 16 does Generalized Additive Models analyses (poisson regression)??

Best wishes,

Cleber.

Eric Zylstra <[hidden email]> escreveu:
Java.

16 has had its interface rewritten in Java. Anything written in Java is much slower than programming code written specifically for a given platform. For me, machines that ran 14 without effort now cycle their fans to high speed under the stress of running Java.

The rewrite was a huge coding job, and I'd be surpised if SPSS was able to do any optimizing. It is hard to program well, especially on a tight release cycle. Add to that writing everything new from scratch. "Admirable" is how I'd word the accomplishment of turning out a functioning yet entirely new interface in the normal release cycle following 15.

This switch will enable SPSS to release one version for all platforms. That efficiency will then be turned into making an even better SPSS for everyone.

Eric

>>> On 5/16/2008 at 9:21 AM, in message
<[hidden email]>, Chris Traynor
wrote:

Eero Olli

Re: SPSS 16.0 lag

In reply to this post by Chris Traynor

Hi Chris and Eric,

I have also noticed the slowness of v16. I have compared runtimes for
one of my longer SPSS scripts (no python, just a production facility
running many plain SPSS syntax files) and on several scripts v.16 spends
eight times more time on a run. (when you compare 30 seconds with 4
minutes, the difference seems like ages).

I am also quite unhappy with the rendering of fonts in jawa based
software (i.e. SPSS v16). On my computer it looks bad as the fonts look
like old pre-GUI fonts. I have checked with SPSS and they say that is
the way it looks in Jawa. It is possible that there is something
uncommon with my hardware and jawa, as I cannot imagine that anyone
would on purpose design a system that looks bad and gives users
headaches.

I have decided to stay on version 15 as long as I can.

Best,
Eero

________________________________________
Eero Olli
Advisor
the Equality and Anti-discrimination Ombud
[hidden email] +47 2405 5951
POB 8048 Dep, N-0031 Oslo, Norway

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

Christian Deindl

statistical question

dear listmembers,

I have a rather simple question

there are two kind of mistakes one could make in test of significance

1) reject the null hypothesis, although it is true (alpha)

2) reject the alternative hypothesis, although it is true (beta)

I have learned that, although the second mistake is lower, if the first
mistake is greater, "beta is not equal 1- alpha".

Now I have found in a rather good book an statistic, that "beta = 1- alpha".

I wonder which statement is true.

does anyone have a ressource or an answer to this question?

thanks a lot,

christian

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

Spousta Jan

Re: statistical question

Hi Christian,

In general, beta is not equal 1- alpha.

But there are cases, where the equation beta = 1 - alpha holds. For example, imagine that you toss a fair coin and I guess (in advance!) whether the outcome will be an eagle. Then my strategy space is limited just by the equation beta = 1 - alpha: I can for example set alpha = 0 and guess "Eagle" in all cases - but then I make an error in all cases where the outcome is not an eagle (beta = 1). Or I can guess "Eagle" in 50% of all cases (alpha = 0.5) and then I achieve beta=0.5. Etc.

But - as you surely see - it means, that in the case beta = 1 - alpha the testing is fairly useless/weak...

Hope this helps,

Jan

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Christian Deindl
Sent: Wednesday, May 21, 2008 1:12 PM
To: [hidden email]
Subject: statistical question

dear listmembers,

I have a rather simple question

there are two kind of mistakes one could make in test of significance

1) reject the null hypothesis, although it is true (alpha)

2) reject the alternative hypothesis, although it is true (beta)

I have learned that, although the second mistake is lower, if the first mistake is greater, "beta is not equal 1- alpha".

Now I have found in a rather good book an statistic, that "beta = 1- alpha".

I wonder which statement is true.

does anyone have a ressource or an answer to this question?

thanks a lot,

christian

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

_____________
Tato zpráva a všechny připojené soubory jsou důvěrné a určené výlučně adresátovi(-ům). Jestliže nejste oprávněným adresátem, je zakázáno jakékoliv zveřejňování, zprostředkování nebo jiné použití těchto informací. Jestliže jste tento mail dostali neoprávněně, prosím, uvědomte odesilatele a smažte zprávu i přiložené soubory. Odesilatel nezodpovídá za jakékoliv chyby nebo opomenutí způsobené tímto přenosem.

Jste si jisti, že opravdu potřebujete vytisknout tuto zprávu a/nebo její přílohy? Myslete na přírodu.

This message and any attached files are confidential and intended solely for the addressee(s). Any publication, transmission or other use of the information by a person or entity other than the intended addressee is prohibited. If you receive this in error please contact the sender and delete the message as well as all attached documents. The sender does not accept liability for any errors or omissions as a result of the transmission.

Are you sure that you really need a print version of this message and/or its attachments? Think about nature.

-.- --

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

Richard Ristow

Re: statistical question

In reply to this post by Christian Deindl

At 07:12 AM 5/21/2008, Christian Deindl wrote:

>there are two kind of mistakes one could make in test of significance
>
>1) reject the null hypothesis, although it is true (alpha)
>2) reject the alternative hypothesis, although it is true (beta)
>
>I have learned that, although the second mistake is lower, if the
>first mistake is greater, "beta is not equal 1- alpha". Now I have
>found in a rather good book an statistic, that "beta = 1- alpha".

As Jan Spousta said, that's not at all true in general.

Analysis of the probability Beta is called "statistical power
analysis". One constant in power analysis is that the information
necessary to determine Alpha does not suffice to determine Beta. For
Beta, you need in addition an 'effect size' -- i.e., *how far* the
null hypothesis is from being true. For example: if you're testing
whether two groups have the same mean, Beta depends on how large the
difference in means actually 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