matrix program runs on 22.0 but not on 21
Locked
 
|
Dear List: I have a matrix program (see bottom on email) that runs correctly when using spss22.0 but when running the program using spss21 I get the following error message(s). The first error message (see below) Error # 12581 A division
by zero has been attempted. Execution of this command stops. It there is a zero somewhere then why does the program run without errors in spss22. Thanks, martin Martin F. Sherman, Ph.D. Professor of Psychology Director of Masters Education in Psychology: Thesis Track Loyola University Maryland Department of Psychology 222 B Beatty Hall 4501 North Charles Street Baltimore, MD 21210 410-617-2417 Run MATRIX procedure: Error encountered in source line # 82 Line 80 compute ww=w*k. Line 81 compute t1=ww*t(v). Line 82. compute procrust=loadings*t1. Error # 12581 A division by zero has been attempted. Execution of this command stops. Error encountered in source line # 82 Error # 12581 A division by zero has been attempted. Error encountered in source line # 83 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'Q1' is undefined Error encountered in source line # 83 Error # 12366 Undefined operand in DIAG. Error encountered in source line # 84 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'K1' is undefined Error encountered in source line # 84 Error # 12586 The argument of the MDIAG function is undefined. Error encountered in source line # 85 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'K' is undefined Error encountered in source line # 85 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'K' is undefined Error encountered in source line # 85 Error # 12343 Undefined operand in matrix multiply. Error encountered in source line # 87 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'WW' is undefined Error encountered in source line # 87 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'WW' is undefined Error encountered in source line # 87 Error # 12343 Undefined operand in matrix multiply. Error encountered in source line # 88 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'T1' is undefined Error encountered in source line # 88 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'T1' is undefined Error encountered in source line # 88 Error # 12343 Undefined operand in matrix multiply. Error encountered in source line # 89 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'PROCRUST' is undefined Error encountered in source line # 89 Error # 12359 Undefined operand for transpose. Error encountered in source line # 95 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CM1M2' is undefined Error encountered in source line # 95 Error # 12366 Undefined operand in DIAG. Error encountered in source line # 96 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'PROCRUST' is undefined Error encountered in source line # 96 Error # 12396 Undefined source operand in one of the CMAX, CMIN, CSSQ, CSUM. Error encountered in source line # 98 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CSUM2M1' is undefined Error encountered in source line # 98 Error # 12346 Undefined operand for unary operator. Error encountered in source line # 100 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CSQRTL1' is undefined Error encountered in source line # 100 Error # 12359 Undefined operand for transpose. Error encountered in source line # 102 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CB' is undefined Error encountered in source line # 102 Error # 12366 Undefined operand in DIAG. Error encountered in source line # 103 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CA' is undefined Error encountered in source line # 103 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'CC' is undefined Error encountered in source line # 103 Error # 12347 Undefined operand for binary operator. Error encountered in source line # 105 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CD' is undefined Error encountered in source line # 105 Error # 12359 Undefined operand for transpose. Error encountered in source line # 106 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'PROCRUST' is undefined Error encountered in source line # 106 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'PROCRUST' is undefined Error encountered in source line # 106 Error # 12343 Undefined operand in matrix multiply. Error encountered in source line # 112 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'RM1M2' is undefined Error encountered in source line # 112 Error # 12366 Undefined operand in DIAG. Error encountered in source line # 113 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'PROCRUST' is undefined Error encountered in source line # 113 Error # 12422 Undefined source operand for one of the RMAX, RMIN, RSSQ, RSUM. Error encountered in source line # 115 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'RSUM2M1' is undefined Error encountered in source line # 115 Error # 12346 Undefined operand for unary operator. Error encountered in source line # 117 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'RSQRTL1' is undefined Error encountered in source line # 117 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'RSQRTL1' is undefined Error encountered in source line # 117 Error # 12343 Undefined operand in matrix multiply. Error encountered in source line # 119 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'RB' is undefined Error encountered in source line # 119 Error # 12366 Undefined operand in DIAG. Error encountered in source line # 120 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'RA' is undefined Error encountered in source line # 120 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'RC' is undefined Error encountered in source line # 120 Error # 12347 Undefined operand for binary operator. Error encountered in source line # 122 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CA' is undefined Error encountered in source line # 122 Error # 12363 Undefined operand in the expression inside brackets. Error encountered in source line # 127 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CC' is undefined Error encountered in source line # 127 Error # 12363 Undefined operand in the expression inside brackets. Error encountered in source line # 128 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'TOP' is undefined Error encountered in source line # 128 Error # 12396 Undefined source operand in one of the CMAX, CMIN, CSSQ, CSUM. Error encountered in source line # 129 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'BOT' is undefined Error encountered in source line # 129 Error # 12396 Undefined source operand in one of the CMAX, CMIN, CSSQ, CSUM. Error encountered in source line # 130 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'CTOP' is undefined Error encountered in source line # 130 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Matrix - 'CBOT' is undefined Error encountered in source line # 130 Error # 12347 Undefined operand for binary operator. Error encountered in source line # 131 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'PROCRUST' is undefined Error encountered in source line # 131 Error # 12363 Undefined operand in the expression inside brackets. Error encountered in source line # 137 Error # 12492 An attempt has been made to use previously undefined matrix (or scalar). Execution of this command stops. Matrix - 'PROCRUST' is undefined Error encountered in source line # 137 Error # 12332 Undefined variable in PRINT. ------ END MATRIX ----- matrix. compute loadings={ -.73, -.26, .03, .15, -.01; -.43, -.33, -.08, -.02, -.60; -.79, -.30, .08, .01, -.11; -.65, -.48, -.09, -.06, -.05; -.78, -.02, -.02, -.12, -.25; -.73, -.37, -.01, -.20, -.12; .21, .73, .33, .24, .14; .15, .81, .13, -.05, -.02; .57, .50, .04, .12, -.34; .25 , .62, .00, .32, -.35; -.11, .48, .48, .28, -.28; .39, .67, .23, .15, .06; -.19, -.10, .71, -.20, .01; .10, .19, .73, -.02, .12; -.21, .05, .68, .27, -.21; .26, .22, .50, -.32, .31; .15, .29, .72, .00, -.09; -.02, .02, .64, .15, .08; .49 , .55, -.02, -.07, .23; .34, -.11, -.24, .25, .63; .19, .39, .33, .49, .41; .27 , .07, .03, -.08, .81; -.25, -.14, .01, -.00, .67; -.15, .28, .24, .65, .18; .66, .26, .10, .51, -.06; .26, -.07, -.06 , .71, -.11; .60, .10, -.04, .60, .11; .62, .38, -.06, .53, -.08; .65, .17, -.02, .55, .00; .75 , -.10, -.06, .17, .11}. compute norms={ .81, .02, -.01, -.01 , -.10; .63, -.03, .01, -.48, -.08; .80, -.10, .02, -.03, -.26; .73, -.18, -.09, .04, -.16; .49, .35, .02 , -.21, -.32; .70, -.15, -.09 , .04, -.38; -.12, .66, .18, .38, .13; -.18, .66, .04, .07, -.03; -.32, .44, .23, -.32, .32; .04, .54, .16, -.27, .42; .00, .58, .11, -.38, -.06; -.04, .74 , .19, .10 , .10; .18, .18 , .58, -.14 , -.31; .14, .04, .73, .17 , .14; .37, .41, .50, -.01 , .12; -.19, .22, .57, .04 , -.04; -.15, -.01, .75, -.09 , .16; -.13, .08, .49, -.07 , -.15; -.35, .22, .15, .56 , .03; -.03, -.15, -.11, .68, .24; -.06, .52, -.05, .55, .27; -.16, -.08, .00, .77, .01; .19, -.12, -.18, .59, -.08; .04, .27, .13 , .62, .00; -.41, .17, .13 , .03, .64; -.04, .06, -.19, .01 , .70; -.20, -.04, .01, .29 , .68; -.09, .23, .15, -.13 , .74; -.33, .17, -.08, .06 , .75; -.23, -.29, -.04 , .22 , .57}. compute s=t(loadings)*norms. compute w1=s*t(s). compute v1=t(s)*s. call eigen(w1,w,evalw1). call eigen(v1,v,evalv1). compute o=t(w)*s*v. compute q1=o &/abs(o). compute k1=diag(q1). compute k=mdiag(k1). compute ww=w*k. compute t1=ww*t(v). compute procrust=loadings*t1. comment the following commands calculate congruence coefficients for the factor columns. compute cm1m2=t(procrust)*norms.
compute ca=diag(cm1m2). compute csum2m1=cssq(procrust).
compute csum2m2=cssq(norms).
compute csqrtl1=sqrt(csum2m1).
compute csqrtl2=sqrt(csum2m2).
compute cb=t(csqrtl1)*csqrtl2.
compute cc=diag(cb).
compute cd=ca&/cc.
compute faccongc=t(cd).
comment the following commands calculate congruence coefficients for the factor rows. compute rm1m2=procrust*t(norms). compute ra=diag(rm1m2). compute rsum2m1=rssq(procrust). compute rsum2m2=rssq(norms). compute rsqrtl1=sqrt(rsum2m1). compute rsqrtl2=sqrt(rsum2m2). compute rb=rsqrtl1*t(rsqrtl2). compute rc=diag(rb). compute faccongr=ra&/rc. comment the following commands calculate the overall congruence coefficients for the data set. compute top={ca;ra}. compute bot={cc;rc}. compute ctop=csum(top). compute cbot=csum(bot). compute total=ctop/cbot. comment the following commands merge the above matrices into one overall matrix that will be printed out. compute procrust={procrust,faccongr;faccongc,total}. print procrust /title = "FACTOR CONGRUENCE COEFFICIENTS"/format f5.2
/clabels= "NEURO" "EXTRA" "OPEN" "AGREE" "CONSC" "ITEMCONG" /RLABELS= "N1" "N2" "N3" "N4" "N5" "N6" "E1" "E2" "E3" "E4" "E5" "E6" "O1" "O2" "O3" "O4" "O5" "O6" "A1" "A2" "A3" "A4" "A5" "A6" "C1" "C2" "C3" "C4" "C5" "C6" "FACTCONG"
/space=newpage.
END MATRIX.
Martin F. Sherman, Ph.D. Professor of Psychology Director of Masters Education in Psychology: Thesis Track Loyola University Maryland Department of Psychology 222 B Beatty Hall 4501 North Charles Street Baltimore, MD 21210 410-617-2417 |
Re: matrix program runs on 22.0 but not on 21
|
There is no division in your lines.
What goes on in line 83? |
Re: matrix program runs on 22.0 but not on 21
Administrator
|
In reply to this post by msherman
Martin,
That program fails in version 22 as well (22.0.0.1 64 bit on Doze8.1). This line is the immediate culprit. compute q1=o &/abs(o). If you inspect the matrix o it shall become obvious why this is the case! Think about/analyze the operation and it should also become clear how to fix it. IMNSHO that is some rather sloppy programming. Since when is it an acceptable practice to perform division without first testing the denominator for 0??? Also, usually the first error in a MATRIX program is sufficient to cause a cascade of errors on anything dependent on the first error. Hence, no real need to post the entire barrage of messages. Thanks for including the code so it was very easy to diagnose. David
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?" |
Re: matrix program runs on 22.0 but not on 21
Administrator
|
In reply to this post by msherman
I have release 21.0.0.2 (64-bit) running under Windoze 7. I tried running your code, and got a bunch of errors, one of them being a division by 0. So I looked at your matrix program to see where the first division is happening. I think it is this line:
compute q1=o &/abs(o). For those not familiar with it, &/ is the element-wise division operator. Next, I inserted a PRINT command to display matrix "o" BEFORE that problematic COMPUTE line, and followed it with an END MATRIX. Here's what o looks like before your matrix program hits that COMPUTE: o-matrix 7.82 .00 .00 .00 .00 .00 -3.76 .00 .00 .00 .00 .00 -2.49 .00 .00 .00 .00 .00 -1.64 .00 .00 .00 .00 .00 1.21 All of the off-diagonal cells are zeros. So that's where your division by 0 is coming from, and I'll bet everything else that goes wrong springs from that. HTH.
--
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/). |
Re: matrix program runs on 22.0 but not on 21
|
Administrator
|
Indeed, that is the line, but you need to format it to see where the 'real' 0 values are.
Note that the result of the &/ operation will be simply the signs of the elements. Run MATRIX procedure: O 7.8238062186462510 -.0000000000000007 -.0000000000000018 .0000000000000000 .0000000000000009 .0000000000000000 -3.7577669793284830 -.0000000000000020 .0000000000000003 -.0000000000000011 -.0000000000000021 .0000000000000009 -2.4867730008316750 -.0000000000000014 -.0000000000000022 .0000000000000022 -.0000000000000014 .0000000000000020 -1.6395375004377320 -.0000000000000007 .0000000000000012 -.0000000000000016 .0000000000000007 -.0000000000000033 1.2102901342792870 ------ END MATRIX -----
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?" |
Re: matrix program runs on 22.0 but not on 21
|
Administrator
|
Furthermore, examining the remainder of the code reveals that ONLY the DIAGonal of the resulting q1 matrix is used in any subsequent calculations. Hence, this full matrix calculation is questionable out of the gate.
A pithy saying comes to mind: "If you think it's expensive to hire a professional, wait until you hire an amateur"
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?" |
Re: matrix program runs on 22.0 but not on 21
|
Administrator
|
This post was updated on .
Odd, Booted my 32 bit box running 22.0.0.1 and the program ran.
Formatting o to 18 decimal places yields the top matrix. This failed in 22.0.0.1 (64 bit with 0 divide error). Failed in 21.?.? with same error (see Bruce W's post). -- Run MATRIX procedure: O 7.823806218646248000 .000000000000001293 .000000000000000238 -.000000000000000805 -.000000000000000157 -.000000000000000875 -3.75776697932848600 .000000000000000952 -.000000000000000096 .000000000000000831 .000000000000000278 -.000000000000000498 -2.48677300083167300 -.000000000000000765 -.000000000000001489 -.000000000000000136 -.000000000000000050 .000000000000000432 -1.63953750043773100 .000000000000000295 -.000000000000000466 .000000000000001078 -.000000000000001309 .000000000000000764 1.210290134279286000 FACTOR CONGRUENCE COEFFICIENTS NEURO EXTRA OPEN AGREE CONSC ITEMCONG N1 .75 -.04 -.09 .02 -.24 .97 N2 .50 -.15 -.11 -.61 -.10 .95 N3 .76 -.08 -.03 -.11 -.36 .99 N4 .65 -.32 -.16 -.09 -.33 .94 N5 .60 .15 -.13 -.26 -.47 .93 N6 .63 -.22 -.08 -.17 -.49 .95 E1 -.28 .71 .28 .25 .19 .96 E2 -.40 .72 .11 .03 -.06 .97 E3 -.53 .42 .10 -.28 .40 .95 E4 -.24 .65 -.02 -.23 .38 .90 E5 .15 .63 .40 -.15 .15 .82 E6 -.45 .59 .23 .14 .24 .84 O1 .21 -.02 .69 .00 -.27 .94 O2 -.05 .22 .72 .16 -.02 .92 O3 .40 .27 .60 -.10 .13 .97 O4 -.38 .09 .55 .27 -.21 .87 O5 -.10 .34 .71 -.03 .05 .89 O6 .17 .13 .60 .15 .09 .73 A1 -.64 .35 .03 .23 .10 .80 A2 -.22 -.23 -.21 .64 .29 .95 A3 -.06 .43 .27 .55 .38 .91 A4 -.36 -.12 .06 .77 -.07 .97 A5 .18 -.16 -.04 .65 -.23 .95 A6 .34 .46 .12 .36 .38 .69 C1 -.37 .24 .14 .07 .75 .99 C2 .14 .06 -.08 .04 .75 .96 C3 -.26 .09 -.01 .24 .78 .98 C4 -.39 .35 -.03 .05 .73 .89 C5 -.33 .15 .02 .12 .78 .99 C6 -.51 -.24 .05 .12 .53 .92 FACTCONG .89 .94 .92 .92 .95 .92 ------ END MATRIX -----
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?" |
| Free forum by Nabble | Edit this page |