PI04896: PROPENSITY SCORE MATCHING FOR JAPANESE DATA DOES NOT WORK WITH UNICODE switched on ON WINDOWS 64 BIT OPERATING SYSTEM

Subscribe to this APAR

By subscribing, you receive periodic emails alerting you to the status of the APAR, along with a link to the fix after it becomes available. You can track this item individually or track all items by product.

Notify me when this APAR changes.

Notify me when an APAR for this component changes.

APAR status

  • Closed as program error.

Error description

  • You work with Statistics 22 on Japanese locale and set unicode
    ON. You run a Propensity Score matching Python syntax. However,
    you notive that the Logistic Regression is not computed.
    Instead you receive this error on output viewer:
    
    Traceback (most recent call last):
      File "<string>", line 21, in <module>
    ValueError: Logistic Regression step failed
    
    With Unicode set to OFF the same syntax runs fine on Japanese
    locale. Please aslo not this happens on Windows 7 64 bit
    operating system. With Windows 32 bit the same syntax works also
    for unicode ON.
    
    This was reported to IBM SPSS development. They revised PSM.spe
    (The custom dialog for Propensity Score Matching). They uploaded
    the revised Version to IBM DeveloperWorks Community. With
    Statistics 22 the installed version of PSM is 1.4.1. The new
    version they created is 1.4.2.
    
    PLEASE NOTE: you cannot use the old PSM syntax you used before.
    You need to install the new PSM bundle and then use the new
    dialog to paste new PSM syntax. This will work for Windows 64
    bit when Unicode is on.
    
    So please do the following:
    - Make sure your computer is connected to the internet and that
    you have an IBM web ID and password (the ID where you normally
    download Fixpacks from FixCentral or download from
    DeveloperWorks)
    - Delete your old PSM syntax file, it will no longer work
    - Launch Statistics 22 and click on Utilities > Extension
    Bundles > Download and Install Extension Bundles
    - A new dialog will occur called Download Extension bundles
    where you will see a list of downloadables
    - Locate PSM (Propensity Score Matching) and select to install
    it: latest version is 1.4.2 and the installed version is 1.4.1
    - Select Download and install selected extension bundles, click
    OK
    - Then you must enter your IBM ID and password and accept the
    Terms of Use, click OK
    - The bundle will be installed and you will be notified that the
    installation was successful. In Output Viewer you will see this
    log:
    
    PSM:Dialog installed to:
    C:\PROGRAM FILES\IBM\SPSS\STATISTICS\22\ext\lib\PSM.
    
    - Next, set locale to Japanese and Unicode ON.
    - Open the files you need for Propensity Score Matching
    - Click Data - Propensity Score Matching
    - Select the variables you used before and define the other
    necessary parameters on this dialog. Click OK and the analysis
    will be computed
    - You can then go back to the dialog and paste the new syntax
    for later usage.
    

Local fix

Problem summary

  • ****************************************************************
    * USERS AFFECTED:                                              *
    * see APAR text                                                *
    ****************************************************************
    * PROBLEM DESCRIPTION:                                         *
    * see APAR text                                                *
    ****************************************************************
    * RECOMMENDATION:                                              *
    * This issue has been addressed in IBM SPSS Statistics 22 Fix  *
    * Pack 1                                                       *
    ****************************************************************
    

Problem conclusion

  • This issue has been addressed in IBM SPSS Statistics 22 Fix Pack
    1
    

Temporary fix

Comments

APAR Information

  • APAR number

    PI04896

  • Reported component name

    SPSS STATISTICS

  • Reported component ID

    5725A54ST

  • Reported release

    M00

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2013-10-28

  • Closed date

    2014-03-19

  • Last modified date

    2014-03-19

  • APAR is sysrouted FROM one or more of the following:

  • APAR is sysrouted TO one or more of the following:

Fix information

  • Fixed component name

    SPSS STATISTICS

  • Fixed component ID

    5725A54ST

Applicable component levels

  • RM00 PSN

       UP

  • RM00 PSY

       UP



Rate this page:

(0 users)Average rating

Add comments

Document information


More support for:

SPSS Statistics
Statistics Desktop

Software version:

22.0

Reference #:

PI04896

Modified date:

2014-03-19

Translate my page

Machine Translation

Content navigation