SPSSX Discussion

Fw: SEEKING HELP WITH MATCH FILES - ONE TO MANY MERGE - REPLY - syntax works-partially

Classic

List

Threaded

6 messages Options

thara vardhan-2

Fw: SEEKING HELP WITH MATCH FILES - ONE TO MANY MERGE - REPLY - syntax works-partially

Hi David

Thanks you so much for helping me with the match file problem. I used the following syntax

MATCH FILES / FILE * / TABLE "unique.sav"/ BY Erefnum increfnum PersCNI .
ADD FILES/FILE * / IN=@M@ /FILE "unique.sav" /IN=@U@/ BY Erefnum increfnum
PersCNI.
COMPUTE
#FLAG=SUM(Erefnum-LAG(Erefnum),increfnum-LAG(increfnum),PersCNI-LAG(PersCNI))
EQ 0 AND LAG(@M@) AND @U@.
SELECT IF NOT #FLAG.
SAVE OUTFILE ' TESTING'.

Got an error message Error # 63 in column 30. Text: unique.sav
The file does not exist.
Execution of this command stops.

So I substituted the filename “unique.sav” with the actual dataset number –
the syntax works fine but I get only 2 new variables as @M@ and @U@
in the new data file testing.sav

MATCH FILES / FILE * / TABLE dataset10/ BY Erefnum increfnum PersCNI .
ADD FILES/FILE * / IN=@M@ /FILE dataset10 /IN=@U@/ BY Erefnum increfnum
PersCNI.
COMPUTE
#FLAG=SUM(Erefnum-LAG(Erefnum),increfnum-LAG(increfnum),PersCNI-LAG(PersCNI))
EQ 0 AND LAG(@M@) AND @U@.
SELECT IF NOT #FLAG.
SAVE OUTFILE ' TESTING'.

Can you help me please?

As you have correctly pointed out I should have outlined the format of the current data files and
also what the format of the output data file that I am looking for.

So here is my request for help again:

UNIQUE.SAV (51330 RECORDS)
ERefNum
IncRefNUm
PersCNI
IncidentCategory
IncidentFurtherClassification
LKAAddress
LKASuburb
LKAPostcode
UniqueParties

MULTI.SAV (624 RECORDS) - there are 24 cases (ErefIncrefPerscni) with duplicate records - These are due to a) multiple law part codes b) a person can have both court and non court actions in one event - that is CHARGE (shown as chargerefno) and an INFRINGEMENT NOTICE(infringement number). c) a young person can be given a charge and youth ConferenceYOA (type of legal action) d) if the a young person has been given a youth Conference YOA - he has 3 or 4 records because law part codes are different.

ERefNum
IncRefNUm
PersCNI
LawPartCode
ChargeRefNo
InfringementNumber
LegalAction
LegalActDate
LegalActMonth
CourtName
CourtType
FirstCourtAppearanceDate
NextCourtAppearanceDate
NextCourtAppearanceType
LawPartTitle
LawPartDescr
CourtResult
ChargeTime24h
CANServedDate
CANServedTime24h
CANs
Outcome
NCA

MATCH RESULT FILE (51330 THAT INCLUDES THE 624) THAT I HAVE TO CREATE

ERefNum
IncRefNUm
PersCNI
IncidentCategory
IncidentFurtherClassification
LKAAddress
LKASuburb
LKAPostcode
UniqueParties
LawPartCode
ChargeRefNo
InfringementNumber
LegalAction
LegalActDate
LegalActMonth
CourtName
CourtType
FirstCourtAppearanceDate
NextCourtAppearanceDate
NextCourtAppearanceType
LawPartTitle
LawPartDescr
CourtResult
ChargeTime24h
CANServedDate
CANServedTime24h
CANs
Outcome
NCA

Many thanks
thara vardhan

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

The information contained in this email is intended for the named recipient(s)
only. It may contain private, confidential, copyright or legally privileged
information. If you are not the intended recipient or you have received this
email by mistake, please reply to the author and delete this email immediately.
You must not copy, print, forward or distribute this email, nor place reliance
on its contents. This email and any attachment have been virus scanned. However,
you are requested to conduct a virus scan as well. No liability is accepted
for any loss or damage resulting from a computer virus, or resulting from a delay
or defect in transmission of this email or any attached file. This email does not
constitute a representation by the NSW Police Force unless the author is legally
entitled to do so.

Mikki Haegle

Automatic reply: SEEKING HELP WITH MATCH FILES - ONE TO MANY MERGE - REPLY - syntax works-partially

I am off enjoying some vacation time.

I will return on June 5th and will respond to messages once I am back in the Lab.

Mikki Haegle

Psychology Lab Coordinator

700 E 7th Street, NM-L202

St Paul, MN 55106

651.793.1354

San K

Re: Fw: SEEKING HELP WITH MATCH FILES - ONE TO MANY MERGE - REPLY - syntax works-partially

In reply to this post by thara vardhan-2

There are few things to note.
1. Make sure your file path is correct for your unique.sav. Generally I simply define the high level path and then refer other files from there.
Eg:
DEFINE !projDir() 'D:\BasixComplianceAnalysis_San\' !ENDDEFINE.
Then you can refer to any other files as
GET FILE = !projDir + ‘analysis\unique.sav’.

This enables me to easily move folders. Also it doesn't depend on the default working dir.

2. Always use ‘EXECUTE’ after using LAG.

3. I generally use ‘EXECUTE’ after a ‘MATCH’ so any errors related to the match is immediately reported.

4. Use commands such as FREQ, CROSSTAB or DESCRIPTIVE every now and then. It will slowdown the process but you can keep an eye on how things are going.

Regards,
San

On Thu, May 24, 2012 at 8:34 AM, Thara Vardhan <[hidden email]> wrote:

Hi David

Thanks you so much for helping me with the match file problem. I used the following syntax

MATCH FILES / FILE * / TABLE "unique.sav"/ BY Erefnum increfnum PersCNI .
ADD FILES/FILE * / IN=@M@ /FILE "unique.sav" /IN=@U@/ BY Erefnum increfnum
PersCNI.
COMPUTE
#FLAG=SUM(Erefnum-LAG(Erefnum),increfnum-LAG(increfnum),PersCNI-LAG(PersCNI))
EQ 0 AND LAG(@M@) AND @U@.
SELECT IF NOT #FLAG.
SAVE OUTFILE ' TESTING'.

Got an error message Error # 63 in column 30. Text: unique.sav
The file does not exist.
Execution of this command stops.

So I substituted the filename “unique.sav” with the actual dataset number –
the syntax works fine but I get only 2 new variables as @M@ and @U@
in the new data file testing.sav

MATCH FILES / FILE * / TABLE dataset10/ BY Erefnum increfnum PersCNI .
ADD FILES/FILE * / IN=@M@ /FILE dataset10 /IN=@U@/ BY Erefnum increfnum
PersCNI.
COMPUTE
#FLAG=SUM(Erefnum-LAG(Erefnum),increfnum-LAG(increfnum),PersCNI-LAG(PersCNI))
EQ 0 AND LAG(@M@) AND @U@.
SELECT IF NOT #FLAG.
SAVE OUTFILE ' TESTING'.

Can you help me please?

As you have correctly pointed out I should have outlined the format of the current data files and
also what the format of the output data file that I am looking for.

So here is my request for help again:

UNIQUE.SAV (51330 RECORDS)
ERefNum
IncRefNUm
PersCNI
IncidentCategory
IncidentFurtherClassification
LKAAddress
LKASuburb
LKAPostcode
UniqueParties

MULTI.SAV (624 RECORDS) - there are 24 cases (ErefIncrefPerscni) with duplicate records - These are due to a) multiple law part codes b) a person can have both court and non court actions in one event - that is CHARGE (shown as chargerefno) and an INFRINGEMENT NOTICE(infringement number). c) a young person can be given a charge and youth ConferenceYOA (type of legal action) d) if the a young person has been given a youth Conference YOA - he has 3 or 4 records because law part codes are different.

ERefNum
IncRefNUm
PersCNI
LawPartCode
ChargeRefNo
InfringementNumber
LegalAction
LegalActDate
LegalActMonth
CourtName
CourtType
FirstCourtAppearanceDate
NextCourtAppearanceDate
NextCourtAppearanceType
LawPartTitle
LawPartDescr
CourtResult
ChargeTime24h
CANServedDate
CANServedTime24h
CANs
Outcome
NCA

MATCH RESULT FILE (51330 THAT INCLUDES THE 624) THAT I HAVE TO CREATE

ERefNum
IncRefNUm
PersCNI
IncidentCategory
IncidentFurtherClassification
LKAAddress
LKASuburb
LKAPostcode
UniqueParties
LawPartCode
ChargeRefNo
InfringementNumber
LegalAction
LegalActDate
LegalActMonth
CourtName
CourtType
FirstCourtAppearanceDate
NextCourtAppearanceDate
NextCourtAppearanceType
LawPartTitle
LawPartDescr
CourtResult
ChargeTime24h
CANServedDate
CANServedTime24h
CANs
Outcome
NCA

Many thanks
thara vardhan

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

The information contained in this email is intended for the named recipient(s)
only. It may contain private, confidential, copyright or legally privileged
information. If you are not the intended recipient or you have received this
email by mistake, please reply to the author and delete this email immediately.
You must not copy, print, forward or distribute this email, nor place reliance
on its contents. This email and any attachment have been virus scanned. However,
you are requested to conduct a virus scan as well. No liability is accepted
for any loss or damage resulting from a computer virus, or resulting from a delay
or defect in transmission of this email or any attached file. This email does not
constitute a representation by the NSW Police Force unless the author is legally
entitled to do so.

David Marso

Re: Fw: SEEKING HELP WITH MATCH FILES - ONE TO MANY MERGE - REPLY - syntax works-partially

Administrator

In reply to this post by thara vardhan-2

Since you have only posted your partial syntax I can only assume that the initial MATCH may have failed or the active file was the FILE=* as well as the TABLE or there was some other error?

Thara Vardhan wrote

Hi David

Thanks you so much for helping me with the match file problem. I used the
following syntax

MATCH FILES / FILE * / TABLE "unique.sav"/ BY Erefnum increfnum PersCNI
.
ADD FILES/FILE * / IN=@M@ /FILE "unique.sav" /IN=@U@/ BY Erefnum
increfnum
PersCNI.
COMPUTE
#FLAG=SUM(Erefnum-LAG(Erefnum),increfnum-LAG(increfnum),PersCNI-LAG(PersCNI))

EQ 0 AND LAG(@M@) AND @U@.
SELECT IF NOT #FLAG.
SAVE OUTFILE ' TESTING'.

Got an error message Error # 63 in column 30. Text: unique.sav
The file does not exist.
Execution of this command stops.

So I substituted the filename ?unique.sav? with the actual dataset number
?
the syntax works fine but I get only 2 new variables as @M@ and @U@
in the new data file testing.sav

MATCH FILES / FILE * / TABLE dataset10/ BY Erefnum increfnum PersCNI .
ADD FILES/FILE * / IN=@M@ /FILE dataset10 /IN=@U@/ BY Erefnum increfnum

PersCNI.
COMPUTE
#FLAG=SUM(Erefnum-LAG(Erefnum),increfnum-LAG(increfnum),PersCNI-LAG(PersCNI))

EQ 0 AND LAG(@M@) AND @U@.
SELECT IF NOT #FLAG.
SAVE OUTFILE ' TESTING'.

Can you help me please?

As you have correctly pointed out I should have outlined the format of the
current data files and
also what the format of the output data file that I am looking for.

So here is my request for help again:

UNIQUE.SAV (51330 RECORDS)
ERefNum
IncRefNUm
PersCNI
IncidentCategory
IncidentFurtherClassification
LKAAddress
LKASuburb
LKAPostcode
UniqueParties

MULTI.SAV (624 RECORDS) - there are 24 cases (ErefIncrefPerscni) with
duplicate records - These are due to a) multiple law part codes b) a
person can have both court and non court actions in one event - that is
CHARGE (shown as chargerefno) and an INFRINGEMENT NOTICE(infringement
number). c) a young person can be given a charge and youth ConferenceYOA
(type of legal action) d) if the a young person has been given a youth
Conference YOA - he has 3 or 4 records because law part codes are
different.

ERefNum
IncRefNUm
PersCNI
LawPartCode
ChargeRefNo
InfringementNumber
LegalAction
LegalActDate
LegalActMonth
CourtName
CourtType
FirstCourtAppearanceDate
NextCourtAppearanceDate
NextCourtAppearanceType
LawPartTitle
LawPartDescr
CourtResult
ChargeTime24h
CANServedDate
CANServedTime24h
CANs
Outcome
NCA

MATCH RESULT FILE (51330 THAT INCLUDES THE 624) THAT I HAVE TO CREATE

ERefNum
IncRefNUm
PersCNI
IncidentCategory
IncidentFurtherClassification
LKAAddress
LKASuburb
LKAPostcode
UniqueParties
LawPartCode
ChargeRefNo
InfringementNumber
LegalAction
LegalActDate
LegalActMonth
CourtName
CourtType
FirstCourtAppearanceDate
NextCourtAppearanceDate
NextCourtAppearanceType
LawPartTitle
LawPartDescr
CourtResult
ChargeTime24h
CANServedDate
CANServedTime24h
CANs
Outcome
NCA

Many thanks
thara vardhan

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

The information contained in this email is intended for the named recipient(s)
only. It may contain private, confidential, copyright or legally privileged
information. If you are not the intended recipient or you have received this
email by mistake, please reply to the author and delete this email immediately.
You must not copy, print, forward or distribute this email, nor place reliance
on its contents. This email and any attachment have been virus scanned. However,
you are requested to conduct a virus scan as well. No liability is accepted
for any loss or damage resulting from a computer virus, or resulting from a delay
or defect in transmission of this email or any attached file. This email does not
constitute a representation by the NSW Police Force unless the author is legally
entitled to do so.

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?"

Jarrod Teo-2

Interaction in Regression Models

Hi,

I have a question to clarify on interaction with Linear Regression.

May I know if the following suggestions are true?

Suggestion 1

The interaction between 2 variables is enough to rule out the relationship.

If no interaction between A&B, then there will not have any interaction between all the ABx, ABxx, ...,ABxxZ.

Suggestion 2

So does this means that if a project is needed to build the following model.

Y=M₁*A+M₂*B+M₃*C+M₄*D+…+ M₁ M₂*A B+ M₁ M₃*AC+ M₁ M₄*AD+…+ M₁ M₂ M₃*ABC+….+M₁M₂…..MnABCDEF…..Z + Constant

Just testing the following will do?

In order to find variables that have no interaction with others w/o using statisitcal model, you got to go to basic method plotting a pair of variables A vs B, A vs C,…, Y vs Z. For any pair that they don’t have interaction, the plot will demonstrate a random pattern.

Hence for a variable that doesn’t have any interaction with others, it must show random pattern in all plots with other variables.

Thanks for any suggestions.

Warmest Regards

Dorraj Oet

David Marso

Re: Interaction in Regression Models

Administrator

Just as it is possible to have non significant main effects but have a sig 2 way interaction, it is possible to have non sig main effects, non sig 2 way effects but a significant 3 way interaction. So, each of your 'suggestions' are demonstrably FALSE!
--

DorraJ Oet wrote

Hi,

I have a question to clarify on interaction with Linear Regression.

May I know if the following suggestions are true?

Suggestion 1

The interaction between 2 variables is enough to rule out the relationship.
If no interaction between A&B, then there will not have any interaction between all the ABx, ABxx, ...,ABxxZ.

Suggestion 2

So does this means that if a project is needed to build the following model.

Y=M1*A+M2*B+M3*C+M4*D+…+ M1 M2*A B+ M1 M3*AC+ M1 M4*AD+…+ M1 M2 M3*ABC+….+M1M2…..MnABCDEF…..Z + Constant

Just testing the following will do?

In order to find variables that have no interaction with others w/o using statisitcal model, you got to go to basic method plotting a pair of variables A vs B, A vs C,…, Y vs Z. For any pair that they don’t have interaction, the plot will demonstrate a random pattern.
Hence for a variable that doesn’t have any interaction with others, it must show random pattern in all plots with other variables.

Thanks for any suggestions.

Warmest Regards
Dorraj Oet