** Mx startup successful ** **MX-Linux version 1.52b** The following MX script lines were read for group 1 G1 CALCULATION GROUP CONTROL GROUP CORRELATION MATRIX A FACTOR DA CALC NG=4 MATRICES A LO 2 2 FIXED C LO 2 2 FREE E ST 2 2 FREE P FULL 4 4 I IDEN 4 4 H FU 1 1 BEGIN ALGEBRA; X = A*A' ; Y = C*C' ; Z = E ; END ALGEBRA; COMPUTE \STND ((I-P)~ *( X+Y+Z| H@X+Y _ H@X+Y| X+Y+Z)*((I-P)~)') / SPECIFY P 0 5 0 0 0 0 0 0 0 0 0 5 0 0 0 0 MATRIX H .5 MATRIX A 0 0 0 MATRIX C 1 0 1 MATRIX E 0 SPECIFY C 1 0 2 SPECIFY E 0 BOUND 0 1 5 OPTION RS SE STANDARD ERRORS withdrawn in this version Please use interval command instead END The following MX script lines were read for group 2 G2 CALCULATION GROUP CLINICAL GROUP CORRELATION MATRIX A FACTOR DA CALC MATRICES A LO 2 2 = A1 C LO 2 2 = C1 E ST 2 2 = E1 P FULL 4 4 = P1 I IDEN 4 4 = I1 H FU 1 1 ! .5 BEGIN ALGEBRA; X = A*A' ; Y = C*C' ; Z = E ; END ALGEBRA; COMPUTE \STND ((I-P)~ *( X+Y+Z| H@X+Y _ H@X+Y| X+Y+Z)*((I-P)~)') / MATRIX H .5 OPTION RS SE STANDARD ERRORS withdrawn in this version Please use interval command instead END The following MX script lines were read for group 3 CORRELATED LIABILITY MODEL: CONTROL GROUP DATA DATA NI=6 ORDINAL FI=CONTROL.DAT Ordinal data read initiated NOTE: Rectangular file contained 422 records with data that contained a total of 2532 observations LABELS AGE1 AGE2 SEX1 SEX2 TYPE DUMMY DEFINITION_VARIABLES AGE1 AGE2 SEX1 SEX2 TYPE / NOTE: Definition yields 422 data vectors for analysis NOTE: Vectors contain a total of 422 observations MATRICES I IDEN 1 1 R FULL 4 4 =%E1 ! CORRELATION MATRIX A1B1A2B2 A FULL 1 4 ! MATRIX FOR AGE DEFINITION VARIABLE S FULL 1 4 ! MATRIX FOR SEX DEFINITION VARIABLE L FULL 1 4 ! ESTIMATED THRESHOLD FOR A O FULL 1 4 ! AGE DIFFERENCE IN THRESHOLD D FULL 1 4 ! SEX DIFFERENCE IN THRESHOLD M FULL 1 1 ! MEAN OF DUMMY VARIABLE V FULL 1 1 ! VARIANCE OF DUMMY VARIABLE J FULL 4 1 ! FOR EXTRACTING RIGHT PART OF K Z ZERO 1 1 BEGIN ALGEBRA ; T = L+(A.O)+(S.D); K = \MULN(R_T_T_(I|I|I|I)) _ ! 1 \MULN(R_T_T_(I|Z|I|I)) _ ! 2 \MULN(R_T_T_(I|I|I|Z)) _ ! 3 \MULN(R_T_T_(Z|I|I|I)) _ ! 4 \MULN(R_T_T_(I|I|Z|I)) _ ! 5 \MULN(R_T_T_(Z|Z|I|I)) _ ! 6 \MULN(R_T_T_(I|I|Z|Z)) _ ! 7 \MULN(R_T_T_(I|Z|I|Z)) _ ! 8 \MULN(R_T_T_(Z|I|I|Z)) _ ! 9 \MULN(R_T_T_(I|Z|Z|I)) _ ! 10 \MULN(R_T_T_(Z|Z|I|Z)) _ ! 11 \MULN(R_T_T_(I|Z|Z|Z)) _ ! 12 \MULN(R_T_T_(Z|I|Z|I)) _ ! 13 \MULN(R_T_T_(Z|Z|Z|I)) _ ! 14 \MULN(R_T_T_(Z|I|Z|Z)) _ ! 15 \MULN(R_T_T_(Z|Z|Z|Z)) ; ! 16 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 END ALGEBRA ; MATRIX J 1 1 1 1 MATRIX M 0 MATRIX V .159154943 MATRIX L 2.0 1.5 2.0 1.5 SPECIFY L 20 21 20 21 ! THRESHOLD FOR A SIB 1; THRESHOLD FOR B SIB 1; THRESHOLD FOR A SIB 2; THRESHOLD FOR B SIB 2; SPECIFY O 30 31 30 31 SPECIFY D 40 41 40 41 SP A AGE1 AGE1 AGE2 AGE2; ! AGE FOR SIB 1; AGE FOR SIB 1; AGE FOR SIB 2; AGE FOR SIB 2; SP S SEX1 SEX1 SEX2 SEX2; ! AGE FOR SIB 1; AGE FOR SIB 1; AGE FOR SIB 2; AGE FOR SIB 2; SP J -5 0 -5 0; MEANS M; COVARIANCE V; WEIGHT (\PART(K,J)) / OPTION RS END The following MX script lines were read for group 4 CORRELATED LIABILITY MODEL: INEFFICIENT APPROACH; CLINICAL GROUP DATA DATA NI=6 ORDINAL FI=CLINICAL.DAT Ordinal data read initiated NOTE: Rectangular file contained 253 records with data that contained a total of 1518 observations LABELS AGE1 AGE2 SEX1 SEX2 TYPE DUMMY DEFINITION_VARIABLES AGE1 AGE2 SEX1 SEX2 TYPE / Note: Global variable previously defined. Updating AGE1 Note: Global variable previously defined. Updating AGE2 Note: Global variable previously defined. Updating SEX1 Note: Global variable previously defined. Updating SEX2 Note: Global variable previously defined. Updating TYPE NOTE: Definition yields 253 data vectors for analysis NOTE: Vectors contain a total of 253 observations MATRICES I IDEN 1 1 R FULL 4 4 =%E2 ! CORRELATION MATRIX A1B1A2B2 A FULL 1 4 ! MATRIX FOR AGE DEFINITION VARIABLE S FULL 1 4 ! MATRIX FOR SEX DEFINITION VARIABLE L FULL 1 4 =L3! ESTIMATED THRESHOLD FOR A O FULL 1 4 =O3! AGE DIFFERENCE IN THRESHOLD D FULL 1 4 =D3! SEX DIFFERENCE IN THRESHOLD M FULL 1 1 ! MEAN OF DUMMY VARIABLE V FULL 1 1 ! VARIANCE OF DUMMY VARIABLE J FULL 4 1 ! FOR EXTRACTING RIGHT PART OF K Z ZERO 1 1 B FULL 1 1 C FULL 1 1 FREE BEGIN ALGEBRA ; G = B + C; T = L+(A.O)+(S.D); K = B.(\MULN(R_T_T_(I|Z|I|I))) _ ! 2 C.(\MULN(R_T_T_(Z|I|I|I))) _ ! 4 G.(\MULN(R_T_T_(Z|Z|I|I))) _ ! 6 B.(\MULN(R_T_T_(I|Z|I|Z))) _ ! 8 C.(\MULN(R_T_T_(Z|I|I|Z))) _ ! 9 B.(\MULN(R_T_T_(I|Z|Z|I))) _ ! 10 G.(\MULN(R_T_T_(Z|Z|I|Z))) _ ! 11 B.(\MULN(R_T_T_(I|Z|Z|Z))) _ ! 12 C.(\MULN(R_T_T_(Z|I|Z|I))) _ ! 13 G.(\MULN(R_T_T_(Z|Z|Z|I))) _ ! 14 C.(\MULN(R_T_T_(Z|I|Z|Z))) _ ! 15 G.(\MULN(R_T_T_(Z|Z|Z|Z))) ; ! 16 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 3 and 1 It is supplied as 1.000000000000000 which has absolute value > .99 Problem with correlation between variables 4 and 2 It is supplied as 1.000000000000000 which has absolute value > .99 Y = \SUM(K); U = Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y; N = (K%U); END ALGEBRA ; MATRIX J 1 1 1 1 MATRIX M 0 MATRIX V .159154943 MATRIX B 1 MATRIX C .5 SP A AGE1 AGE1 AGE2 AGE2; ! AGE FOR SIB 1; AGE FOR SIB 1; AGE FOR SIB 2; AGE FOR SIB 2; SP S SEX1 SEX1 SEX2 SEX2; ! AGE FOR SIB 1; AGE FOR SIB 1; AGE FOR SIB 2; AGE FOR SIB 2; SP J -5 0 -5 0; MEANS M; COVARIANCE V; WEIGHT (\PART(N,J)) / OPTION FUNC=.000001 OPTION RS OPTION TH=-5 END Summary of VL file data for group 3 TYPE SEX2 SEX1 AGE2 AGE1 Code -5.0000E+00 -4.0000E+00 -3.0000E+00 -2.0000E+00 -1.0000E+00 Number 4.2200E+02 4.2200E+02 4.2200E+02 4.2200E+02 4.2200E+02 Mean 1.6303E+00 5.4976E-01 9.1943E-01 1.6690E+01 1.6478E+01 Variance 3.8017E+00 2.4752E-01 7.4077E-02 1.1376E+01 2.2599E+00 Minimum 1.0000E+00 0.0000E+00 0.0000E+00 1.0960E+01 1.1960E+01 Maximum 1.6000E+01 1.0000E+00 1.0000E+00 2.4950E+01 2.1050E+01 DUMMY Code 1.0000E+00 Number 4.2200E+02 Mean 1.0000E+00 Variance 2.6940E-16 Minimum 1.0000E+00 Maximum 1.0000E+00 Summary of VL file data for group 4 TYPE SEX2 SEX1 AGE2 AGE1 DUMMY Code -5.0000 -4.0000 -3.0000 -2.0000 -1.0000 1.0000 Number 253.0000 253.0000 253.0000 253.0000 253.0000 253.0000 Mean 3.1779 0.5375 0.9289 17.0966 16.4344 1.0000 Variance 7.7826 0.2486 0.0661 10.7495 1.4494 0.0000 Minimum 1.0000 0.0000 0.0000 11.1500 13.0900 1.0000 Maximum 12.0000 1.0000 1.0000 24.5500 19.7100 1.0000 PARAMETER SPECIFICATIONS GROUP NUMBER: 1 G1 CALCULATION GROUP Control group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 It has no free parameters specified MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 1 2 0 2 MATRIX E This is a STANDARDISED matrix of order 2 by 2 It has no free parameters specified MATRIX H This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0 5 0 0 2 0 0 0 0 3 0 0 0 5 4 0 0 0 0 MATRIX X This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Y This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Z This is a computed FULL matrix of order 2 by 2 It has no free parameters specified GROUP NUMBER: 2 G2 CALCULATION GROUP Clinical group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 It has no free parameters specified MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 1 2 0 2 MATRIX E This is a STANDARDISED matrix of order 2 by 2 It has no free parameters specified MATRIX H This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0 5 0 0 2 0 0 0 0 3 0 0 0 5 4 0 0 0 0 MATRIX X This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Y This is a computed FULL matrix of order 2 by 2 It has no free parameters specified MATRIX Z This is a computed FULL matrix of order 2 by 2 It has no free parameters specified GROUP NUMBER: 3 Correlated liability model: control group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 -1 -1 -2 -2 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 40 41 40 41 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 -5 2 0 3 -5 4 0 MATRIX K This is a computed FULL matrix of order 16 by 1 It has no free parameters specified MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 20 21 20 21 MATRIX M This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 30 31 30 31 MATRIX R This is a constrained FULL matrix of order 4 by 4 It has no free parameters specified MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 -3 -3 -4 -4 MATRIX T This is a computed FULL matrix of order 1 by 4 It has no free parameters specified MATRIX V This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX Z This is a NULL matrix of order 1 by 1 GROUP NUMBER: 4 Correlated liability model: inefficient approach; clinical group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 -1 -1 -2 -2 MATRIX B This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX C This is a FULL matrix of order 1 by 1 1 1 42 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 40 41 40 41 MATRIX G This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 -5 2 0 3 -5 4 0 MATRIX K This is a computed FULL matrix of order 12 by 1 It has no free parameters specified MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 20 21 20 21 MATRIX M This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX N This is a computed FULL matrix of order 12 by 1 It has no free parameters specified MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 30 31 30 31 MATRIX R This is a constrained FULL matrix of order 4 by 4 It has no free parameters specified MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 -3 -3 -4 -4 MATRIX T This is a computed FULL matrix of order 1 by 4 It has no free parameters specified MATRIX U This is a computed FULL matrix of order 12 by 1 It has no free parameters specified MATRIX V This is a FULL matrix of order 1 by 1 It has no free parameters specified MATRIX Y This is a computed FULL matrix of order 1 by 1 It has no free parameters specified MATRIX Z This is a NULL matrix of order 1 by 1 Mx starting optimization; number of parameters = 10 *** WARNING! *** I am not sure I have found a solution that satisfies Kuhn-Tucker conditions for a minimum. NAG's IFAIL parameter is 6 Looks like I got stuck here. Check the following: 1. The model is correctly specified 2. Starting values are good 3. You are not already at the solution The error can arise if the Hessian is ill-conditioned You can try resetting it to an identity matrix and fit from the solution by putting TH=-n on the OU line where n is the number of refits that you want to do If all else fails try putting NAG=30 on the OU line and examine the file NAGDUMP.OUT and the NAG manual MX PARAMETER ESTIMATES GROUP NUMBER: 1 G1 CALCULATION GROUP Control group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.4526 2 0.0000 0.6393 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.2048 0.0000 2 0.0000 0.4087 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7342 0.2348 0.2130 2 0.7342 1.0000 0.2130 0.2901 3 0.2348 0.2130 1.0000 0.7342 4 0.2130 0.2901 0.7342 1.0000 GROUP NUMBER: 2 G2 CALCULATION GROUP Clinical group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.4526 2 0.0000 0.6393 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.2048 0.0000 2 0.0000 0.4087 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7342 0.2348 0.2130 2 0.7342 1.0000 0.2130 0.2901 3 0.2348 0.2130 1.0000 0.7342 4 0.2130 0.2901 0.7342 1.0000 GROUP NUMBER: 3 Correlated liability model: control group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 12.8900 12.8900 13.9100 13.9100 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.2319 -0.1087 -0.2319 -0.1087 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 16 by 1 [=\MULN(R_T_T_(I|I|I|I))_\MULN(R_T_T_(I|Z|I|I))_\MULN(R_T_T_(I|I|I|Z))_\MULN(R_T_T_(Z|I|I|I))_\MULN(R_T_T_(I|I|Z|I))_\MULN(R_T_T_(Z|Z|I|I))_\MULN(R_T_T_(I|I|Z|Z))_\MULN(R_T_T_(I|Z|I|Z))_\MULN(R_T_T_(Z|I|I|Z))_\MULN(R_T_T_(I|Z|Z|I))_\MULN(R_T_T_(Z|Z|I|Z))_\MULN(R_T_T_(I|Z|Z|Z))_\MULN(R_T_T_(Z|I|Z|I))_\MULN(R_T_T_(Z|Z|Z|I))_\MULN(R_T_T_(Z|I|Z|Z))_\MULN(R_T_T_(Z|Z|Z|Z))] 1 1 9.2988E-01 2 1.4334E-02 3 2.1991E-02 4 4.3364E-03 5 1.2349E-02 6 3.5291E-03 7 9.5894E-03 8 1.3546E-03 9 1.9716E-04 10 3.6872E-04 11 3.3585E-04 12 6.2786E-04 13 1.9712E-04 14 1.6297E-04 15 1.6314E-04 16 2.3683E-04 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1849 3.3916 4.1849 3.3916 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1416 -0.1053 -0.1416 -0.1053 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7342 0.2348 0.2130 2 0.7342 1.0000 0.2130 0.2901 3 0.2348 0.2130 1.0000 0.7342 4 0.2130 0.2901 0.7342 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 0.0000 0.0000 1.0000 1.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 2.3591 2.0345 1.9828 1.8185 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1229.4790 Where the fit function is -2 * Log-likelihood of raw ordinal GROUP NUMBER: 4 Correlated liability model: inefficient approach; clinical group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 15.0800 15.0800 18.1900 18.1900 MATRIX B This is a FULL matrix of order 1 by 1 1 1 1.0000 MATRIX C This is a FULL matrix of order 1 by 1 1 1 0.1649 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.2319 -0.1087 -0.2319 -0.1087 MATRIX G This is a computed FULL matrix of order 1 by 1 [=B+C] 1 1 1.1649 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 12 by 1 [=B.(\MULN(R_T_T_(I|Z|I|I)))_C.(\MULN(R_T_T_(Z|I|I|I)))_G.(\MULN(R_T_T_(Z|Z|I|I)))_B.(\MULN(R_T_T_(I|Z|I|Z)))_C.(\MULN(R_T_T_(Z|I|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|I)))_G.(\MULN(R_T_T_(Z|Z|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|Z)))_C.(\MULN(R_T_T_(Z|I|Z|I)))_G.(\MULN(R_T_T_(Z|Z|Z|I)))_C.(\MULN(R_T_T_(Z|I|Z|Z)))_G.(\MULN(R_T_T_(Z|Z|Z|Z)))] 1 1 2.3037E-02 2 2.5607E-03 3 1.3791E-02 4 3.1903E-03 5 1.7339E-04 6 9.8583E-04 7 1.8853E-03 8 1.9919E-03 9 1.8356E-04 10 9.9423E-04 11 1.8309E-04 12 1.7414E-03 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1849 3.3916 4.1849 3.3916 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX N This is a computed FULL matrix of order 12 by 1 [=(K%U)] 1 1 0.4542 2 0.0505 3 0.2719 4 0.0629 5 0.0034 6 0.0194 7 0.0372 8 0.0393 9 0.0036 10 0.0196 11 0.0036 12 0.0343 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1416 -0.1053 -0.1416 -0.1053 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7342 0.2348 0.2130 2 0.7342 1.0000 0.2130 0.2901 3 0.2348 0.2130 1.0000 0.7342 4 0.2130 0.2901 0.7342 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 1.0000 1.0000 0.0000 0.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 1.8170 1.6953 1.6084 1.4765 MATRIX U This is a computed FULL matrix of order 12 by 1 [=Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y] 1 1 0.0507 2 0.0507 3 0.0507 4 0.0507 5 0.0507 6 0.0507 7 0.0507 8 0.0507 9 0.0507 10 0.0507 11 0.0507 12 0.0507 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Y This is a computed FULL matrix of order 1 by 1 [=\SUM(K)] 1 1 0.0507 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1159.2670 Where the fit function is -2 * Log-likelihood of raw ordinal *** WARNING! *** Minimization may not be successful. See above CODE RED - Hessian/precision problem Your model has 10 estimated parameters and 675 Observed statistics -2 times log-likelihood of data >>> 2388.746 Degrees of freedom >>>>>>>>>>>>>>>> 665 Now I'm trying to improve on the current solution for you... Mx starting optimization; number of parameters = 10 *** WARNING! *** I am not sure I have found a solution that satisfies Kuhn-Tucker conditions for a minimum. NAG's IFAIL parameter is 6 Looks like I got stuck here. Check the following: 1. The model is correctly specified 2. Starting values are good 3. You are not already at the solution The error can arise if the Hessian is ill-conditioned You can try resetting it to an identity matrix and fit from the solution by putting TH=-n on the OU line where n is the number of refits that you want to do If all else fails try putting NAG=30 on the OU line and examine the file NAGDUMP.OUT and the NAG manual MX PARAMETER ESTIMATES GROUP NUMBER: 1 G1 CALCULATION GROUP Control group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 2 G2 CALCULATION GROUP Clinical group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 3 Correlated liability model: control group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 12.8900 12.8900 13.9100 13.9100 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 16 by 1 [=\MULN(R_T_T_(I|I|I|I))_\MULN(R_T_T_(I|Z|I|I))_\MULN(R_T_T_(I|I|I|Z))_\MULN(R_T_T_(Z|I|I|I))_\MULN(R_T_T_(I|I|Z|I))_\MULN(R_T_T_(Z|Z|I|I))_\MULN(R_T_T_(I|I|Z|Z))_\MULN(R_T_T_(I|Z|I|Z))_\MULN(R_T_T_(Z|I|I|Z))_\MULN(R_T_T_(I|Z|Z|I))_\MULN(R_T_T_(Z|Z|I|Z))_\MULN(R_T_T_(I|Z|Z|Z))_\MULN(R_T_T_(Z|I|Z|I))_\MULN(R_T_T_(Z|Z|Z|I))_\MULN(R_T_T_(Z|I|Z|Z))_\MULN(R_T_T_(Z|Z|Z|Z))] 1 1 9.3261E-01 2 1.2104E-02 3 2.2960E-02 4 3.1775E-03 5 1.1995E-02 6 2.8000E-03 7 1.0528E-02 8 1.3510E-03 9 1.7945E-04 10 3.4530E-04 11 3.3085E-04 12 6.9580E-04 13 1.2504E-04 14 1.2354E-04 15 1.3442E-04 16 2.3096E-04 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 0.0000 0.0000 1.0000 1.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 2.4521 2.0974 1.9742 1.7939 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1228.8525 Where the fit function is -2 * Log-likelihood of raw ordinal GROUP NUMBER: 4 Correlated liability model: inefficient approach; clinical group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 15.0800 15.0800 18.1900 18.1900 MATRIX B This is a FULL matrix of order 1 by 1 1 1 1.0000 MATRIX C This is a FULL matrix of order 1 by 1 1 1 0.1701 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX G This is a computed FULL matrix of order 1 by 1 [=B+C] 1 1 1.1701 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 12 by 1 [=B.(\MULN(R_T_T_(I|Z|I|I)))_C.(\MULN(R_T_T_(Z|I|I|I)))_G.(\MULN(R_T_T_(Z|Z|I|I)))_B.(\MULN(R_T_T_(I|Z|I|Z)))_C.(\MULN(R_T_T_(Z|I|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|I)))_G.(\MULN(R_T_T_(Z|Z|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|Z)))_C.(\MULN(R_T_T_(Z|I|Z|I)))_G.(\MULN(R_T_T_(Z|Z|Z|I)))_C.(\MULN(R_T_T_(Z|I|Z|Z)))_G.(\MULN(R_T_T_(Z|Z|Z|Z)))] 1 1 2.4297E-02 2 2.5652E-03 3 1.5169E-02 4 3.2190E-03 5 1.7381E-04 6 8.2548E-04 7 2.1240E-03 8 1.8244E-03 9 1.1596E-04 10 7.7018E-04 11 1.3791E-04 12 1.6010E-03 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX N This is a computed FULL matrix of order 12 by 1 [=(K%U)] 1 1 0.4600 2 0.0486 3 0.2872 4 0.0609 5 0.0033 6 0.0156 7 0.0402 8 0.0345 9 0.0022 10 0.0146 11 0.0026 12 0.0303 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 1.0000 1.0000 0.0000 0.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 1.8197 1.6749 1.7523 1.5583 MATRIX U This is a computed FULL matrix of order 12 by 1 [=Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y] 1 1 0.0528 2 0.0528 3 0.0528 4 0.0528 5 0.0528 6 0.0528 7 0.0528 8 0.0528 9 0.0528 10 0.0528 11 0.0528 12 0.0528 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Y This is a computed FULL matrix of order 1 by 1 [=\SUM(K)] 1 1 0.0528 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1157.2752 Where the fit function is -2 * Log-likelihood of raw ordinal *** WARNING! *** Minimization may not be successful. See above CODE RED - Hessian/precision problem Your model has 10 estimated parameters and 675 Observed statistics -2 times log-likelihood of data >>> 2386.128 Degrees of freedom >>>>>>>>>>>>>>>> 665 Now I'm trying to improve on the current solution for you... Mx starting optimization; number of parameters = 10 MX PARAMETER ESTIMATES GROUP NUMBER: 1 G1 CALCULATION GROUP Control group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 2 G2 CALCULATION GROUP Clinical group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 3 Correlated liability model: control group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 12.8900 12.8900 13.9100 13.9100 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 16 by 1 [=\MULN(R_T_T_(I|I|I|I))_\MULN(R_T_T_(I|Z|I|I))_\MULN(R_T_T_(I|I|I|Z))_\MULN(R_T_T_(Z|I|I|I))_\MULN(R_T_T_(I|I|Z|I))_\MULN(R_T_T_(Z|Z|I|I))_\MULN(R_T_T_(I|I|Z|Z))_\MULN(R_T_T_(I|Z|I|Z))_\MULN(R_T_T_(Z|I|I|Z))_\MULN(R_T_T_(I|Z|Z|I))_\MULN(R_T_T_(Z|Z|I|Z))_\MULN(R_T_T_(I|Z|Z|Z))_\MULN(R_T_T_(Z|I|Z|I))_\MULN(R_T_T_(Z|Z|Z|I))_\MULN(R_T_T_(Z|I|Z|Z))_\MULN(R_T_T_(Z|Z|Z|Z))] 1 1 9.3261E-01 2 1.2118E-02 3 2.2991E-02 4 3.1705E-03 5 1.1969E-02 6 2.7982E-03 7 1.0523E-02 8 1.3539E-03 9 1.7929E-04 10 3.4496E-04 11 3.3103E-04 12 6.9617E-04 13 1.2454E-04 14 1.2322E-04 15 1.3408E-04 16 2.3072E-04 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 0.0000 0.0000 1.0000 1.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 2.4526 2.0970 1.9748 1.7936 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1228.8353 Where the fit function is -2 * Log-likelihood of raw ordinal GROUP NUMBER: 4 Correlated liability model: inefficient approach; clinical group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 15.0800 15.0800 18.1900 18.1900 MATRIX B This is a FULL matrix of order 1 by 1 1 1 1.0000 MATRIX C This is a FULL matrix of order 1 by 1 1 1 0.1701 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX G This is a computed FULL matrix of order 1 by 1 [=B+C] 1 1 1.1701 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 12 by 1 [=B.(\MULN(R_T_T_(I|Z|I|I)))_C.(\MULN(R_T_T_(Z|I|I|I)))_G.(\MULN(R_T_T_(Z|Z|I|I)))_B.(\MULN(R_T_T_(I|Z|I|Z)))_C.(\MULN(R_T_T_(Z|I|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|I)))_G.(\MULN(R_T_T_(Z|Z|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|Z)))_C.(\MULN(R_T_T_(Z|I|Z|I)))_G.(\MULN(R_T_T_(Z|Z|Z|I)))_C.(\MULN(R_T_T_(Z|I|Z|Z)))_G.(\MULN(R_T_T_(Z|Z|Z|Z)))] 1 1 2.4333E-02 2 2.5594E-03 3 1.5162E-02 4 3.2282E-03 5 1.7372E-04 6 8.2443E-04 7 2.1264E-03 8 1.8257E-03 9 1.1542E-04 10 7.6787E-04 11 1.3754E-04 12 1.5994E-03 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX N This is a computed FULL matrix of order 12 by 1 [=(K%U)] 1 1 0.4604 2 0.0484 3 0.2869 4 0.0611 5 0.0033 6 0.0156 7 0.0402 8 0.0345 9 0.0022 10 0.0145 11 0.0026 12 0.0303 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 1.0000 1.0000 0.0000 0.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 1.8203 1.6745 1.7530 1.5578 MATRIX U This is a computed FULL matrix of order 12 by 1 [=Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y] 1 1 0.0529 2 0.0529 3 0.0529 4 0.0529 5 0.0529 6 0.0529 7 0.0529 8 0.0529 9 0.0529 10 0.0529 11 0.0529 12 0.0529 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Y This is a computed FULL matrix of order 1 by 1 [=\SUM(K)] 1 1 0.0529 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1157.2917 Where the fit function is -2 * Log-likelihood of raw ordinal Your model has 10 estimated parameters and 675 Observed statistics -2 times log-likelihood of data >>> 2386.127 Degrees of freedom >>>>>>>>>>>>>>>> 665 Now I'm trying to improve on the current solution for you... Mx starting optimization; number of parameters = 10 *** WARNING! *** I am not sure I have found a solution that satisfies Kuhn-Tucker conditions for a minimum. NAG's IFAIL parameter is 1 We probably have a minimum here, but you might consider trying different starting values. You can randomize these with TH=n on the OU line, where n is the number of times you wish to do this. I STRONGLY recommend BOundaries to be set if you use TH MX PARAMETER ESTIMATES GROUP NUMBER: 1 G1 CALCULATION GROUP Control group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 2 G2 CALCULATION GROUP Clinical group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 3 Correlated liability model: control group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 12.8900 12.8900 13.9100 13.9100 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 16 by 1 [=\MULN(R_T_T_(I|I|I|I))_\MULN(R_T_T_(I|Z|I|I))_\MULN(R_T_T_(I|I|I|Z))_\MULN(R_T_T_(Z|I|I|I))_\MULN(R_T_T_(I|I|Z|I))_\MULN(R_T_T_(Z|Z|I|I))_\MULN(R_T_T_(I|I|Z|Z))_\MULN(R_T_T_(I|Z|I|Z))_\MULN(R_T_T_(Z|I|I|Z))_\MULN(R_T_T_(I|Z|Z|I))_\MULN(R_T_T_(Z|Z|I|Z))_\MULN(R_T_T_(I|Z|Z|Z))_\MULN(R_T_T_(Z|I|Z|I))_\MULN(R_T_T_(Z|Z|Z|I))_\MULN(R_T_T_(Z|I|Z|Z))_\MULN(R_T_T_(Z|Z|Z|Z))] 1 1 9.3261E-01 2 1.2118E-02 3 2.2991E-02 4 3.1705E-03 5 1.1969E-02 6 2.7982E-03 7 1.0523E-02 8 1.3539E-03 9 1.7929E-04 10 3.4496E-04 11 3.3103E-04 12 6.9617E-04 13 1.2454E-04 14 1.2322E-04 15 1.3408E-04 16 2.3072E-04 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 0.0000 0.0000 1.0000 1.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 2.4526 2.0970 1.9748 1.7936 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1228.8353 Where the fit function is -2 * Log-likelihood of raw ordinal GROUP NUMBER: 4 Correlated liability model: inefficient approach; clinical group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 15.0800 15.0800 18.1900 18.1900 MATRIX B This is a FULL matrix of order 1 by 1 1 1 1.0000 MATRIX C This is a FULL matrix of order 1 by 1 1 1 0.1701 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX G This is a computed FULL matrix of order 1 by 1 [=B+C] 1 1 1.1701 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 12 by 1 [=B.(\MULN(R_T_T_(I|Z|I|I)))_C.(\MULN(R_T_T_(Z|I|I|I)))_G.(\MULN(R_T_T_(Z|Z|I|I)))_B.(\MULN(R_T_T_(I|Z|I|Z)))_C.(\MULN(R_T_T_(Z|I|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|I)))_G.(\MULN(R_T_T_(Z|Z|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|Z)))_C.(\MULN(R_T_T_(Z|I|Z|I)))_G.(\MULN(R_T_T_(Z|Z|Z|I)))_C.(\MULN(R_T_T_(Z|I|Z|Z)))_G.(\MULN(R_T_T_(Z|Z|Z|Z)))] 1 1 2.4333E-02 2 2.5594E-03 3 1.5162E-02 4 3.2282E-03 5 1.7372E-04 6 8.2443E-04 7 2.1264E-03 8 1.8257E-03 9 1.1542E-04 10 7.6787E-04 11 1.3754E-04 12 1.5994E-03 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX N This is a computed FULL matrix of order 12 by 1 [=(K%U)] 1 1 0.4604 2 0.0484 3 0.2869 4 0.0611 5 0.0033 6 0.0156 7 0.0402 8 0.0345 9 0.0022 10 0.0145 11 0.0026 12 0.0303 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 1.0000 1.0000 0.0000 0.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 1.8203 1.6745 1.7530 1.5578 MATRIX U This is a computed FULL matrix of order 12 by 1 [=Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y] 1 1 0.0529 2 0.0529 3 0.0529 4 0.0529 5 0.0529 6 0.0529 7 0.0529 8 0.0529 9 0.0529 10 0.0529 11 0.0529 12 0.0529 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Y This is a computed FULL matrix of order 1 by 1 [=\SUM(K)] 1 1 0.0529 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1157.2917 Where the fit function is -2 * Log-likelihood of raw ordinal *** WARNING! *** Minimization may not be successful. See above CODE GREEN - it probably was OK Your model has 10 estimated parameters and 675 Observed statistics -2 times log-likelihood of data >>> 2386.127 Degrees of freedom >>>>>>>>>>>>>>>> 665 Now I'm trying to improve on the current solution for you... Mx starting optimization; number of parameters = 10 *** WARNING! *** I am not sure I have found a solution that satisfies Kuhn-Tucker conditions for a minimum. NAG's IFAIL parameter is 1 We probably have a minimum here, but you might consider trying different starting values. You can randomize these with TH=n on the OU line, where n is the number of times you wish to do this. I STRONGLY recommend BOundaries to be set if you use TH MX PARAMETER ESTIMATES GROUP NUMBER: 1 G1 CALCULATION GROUP Control group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 2 G2 CALCULATION GROUP Clinical group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 3 Correlated liability model: control group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 12.8900 12.8900 13.9100 13.9100 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 16 by 1 [=\MULN(R_T_T_(I|I|I|I))_\MULN(R_T_T_(I|Z|I|I))_\MULN(R_T_T_(I|I|I|Z))_\MULN(R_T_T_(Z|I|I|I))_\MULN(R_T_T_(I|I|Z|I))_\MULN(R_T_T_(Z|Z|I|I))_\MULN(R_T_T_(I|I|Z|Z))_\MULN(R_T_T_(I|Z|I|Z))_\MULN(R_T_T_(Z|I|I|Z))_\MULN(R_T_T_(I|Z|Z|I))_\MULN(R_T_T_(Z|Z|I|Z))_\MULN(R_T_T_(I|Z|Z|Z))_\MULN(R_T_T_(Z|I|Z|I))_\MULN(R_T_T_(Z|Z|Z|I))_\MULN(R_T_T_(Z|I|Z|Z))_\MULN(R_T_T_(Z|Z|Z|Z))] 1 1 9.3261E-01 2 1.2118E-02 3 2.2991E-02 4 3.1705E-03 5 1.1969E-02 6 2.7982E-03 7 1.0523E-02 8 1.3539E-03 9 1.7929E-04 10 3.4496E-04 11 3.3103E-04 12 6.9617E-04 13 1.2454E-04 14 1.2322E-04 15 1.3408E-04 16 2.3072E-04 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 0.0000 0.0000 1.0000 1.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 2.4526 2.0970 1.9748 1.7936 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1228.8353 Where the fit function is -2 * Log-likelihood of raw ordinal GROUP NUMBER: 4 Correlated liability model: inefficient approach; clinical group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 15.0800 15.0800 18.1900 18.1900 MATRIX B This is a FULL matrix of order 1 by 1 1 1 1.0000 MATRIX C This is a FULL matrix of order 1 by 1 1 1 0.1701 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX G This is a computed FULL matrix of order 1 by 1 [=B+C] 1 1 1.1701 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 12 by 1 [=B.(\MULN(R_T_T_(I|Z|I|I)))_C.(\MULN(R_T_T_(Z|I|I|I)))_G.(\MULN(R_T_T_(Z|Z|I|I)))_B.(\MULN(R_T_T_(I|Z|I|Z)))_C.(\MULN(R_T_T_(Z|I|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|I)))_G.(\MULN(R_T_T_(Z|Z|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|Z)))_C.(\MULN(R_T_T_(Z|I|Z|I)))_G.(\MULN(R_T_T_(Z|Z|Z|I)))_C.(\MULN(R_T_T_(Z|I|Z|Z)))_G.(\MULN(R_T_T_(Z|Z|Z|Z)))] 1 1 2.4333E-02 2 2.5594E-03 3 1.5162E-02 4 3.2282E-03 5 1.7372E-04 6 8.2443E-04 7 2.1264E-03 8 1.8257E-03 9 1.1542E-04 10 7.6787E-04 11 1.3754E-04 12 1.5994E-03 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX N This is a computed FULL matrix of order 12 by 1 [=(K%U)] 1 1 0.4604 2 0.0484 3 0.2869 4 0.0611 5 0.0033 6 0.0156 7 0.0402 8 0.0345 9 0.0022 10 0.0145 11 0.0026 12 0.0303 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 1.0000 1.0000 0.0000 0.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 1.8203 1.6745 1.7530 1.5578 MATRIX U This is a computed FULL matrix of order 12 by 1 [=Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y] 1 1 0.0529 2 0.0529 3 0.0529 4 0.0529 5 0.0529 6 0.0529 7 0.0529 8 0.0529 9 0.0529 10 0.0529 11 0.0529 12 0.0529 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Y This is a computed FULL matrix of order 1 by 1 [=\SUM(K)] 1 1 0.0529 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1157.2917 Where the fit function is -2 * Log-likelihood of raw ordinal *** WARNING! *** Minimization may not be successful. See above CODE GREEN - it probably was OK Your model has 10 estimated parameters and 675 Observed statistics -2 times log-likelihood of data >>> 2386.127 Degrees of freedom >>>>>>>>>>>>>>>> 665 Now I'm trying to improve on the current solution for you... Mx starting optimization; number of parameters = 10 *** WARNING! *** I am not sure I have found a solution that satisfies Kuhn-Tucker conditions for a minimum. NAG's IFAIL parameter is 1 We probably have a minimum here, but you might consider trying different starting values. You can randomize these with TH=n on the OU line, where n is the number of times you wish to do this. I STRONGLY recommend BOundaries to be set if you use TH MX PARAMETER ESTIMATES GROUP NUMBER: 1 G1 CALCULATION GROUP Control group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 2 G2 CALCULATION GROUP Clinical group Correlation matrix A factor MATRIX A This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 0.0000 2 0.0000 0.0000 MATRIX C This is a LOWER TRIANGULAR matrix of order 2 by 2 1 2 1 -0.3686 2 0.0000 0.6848 MATRIX E This is a STANDARDISED matrix of order 2 by 2 1 2 1 1.0000 2 0.0000 1.0000 MATRIX H This is a FULL matrix of order 1 by 1 1 1 0.5000 MATRIX I This is an IDENTITY matrix of order 4 by 4 MATRIX P This is a FULL matrix of order 4 by 4 1 2 3 4 1 0.0000 1.0000 0.0000 0.0000 2 0.0000 0.0000 0.0000 0.0000 3 0.0000 0.0000 0.0000 1.0000 4 0.0000 0.0000 0.0000 0.0000 MATRIX X This is a computed FULL matrix of order 2 by 2 [=A*A'] 1 2 1 0.0000 0.0000 2 0.0000 0.0000 MATRIX Y This is a computed FULL matrix of order 2 by 2 [=C*C'] 1 2 1 0.1359 0.0000 2 0.0000 0.4689 MATRIX Z This is a computed FULL matrix of order 2 by 2 [=E] 1 2 1 1.0000 0.0000 2 0.0000 1.0000 EXPECTED MATRIX of this CALCULATION group 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 GROUP NUMBER: 3 Correlated liability model: control group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 12.8900 12.8900 13.9100 13.9100 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 16 by 1 [=\MULN(R_T_T_(I|I|I|I))_\MULN(R_T_T_(I|Z|I|I))_\MULN(R_T_T_(I|I|I|Z))_\MULN(R_T_T_(Z|I|I|I))_\MULN(R_T_T_(I|I|Z|I))_\MULN(R_T_T_(Z|Z|I|I))_\MULN(R_T_T_(I|I|Z|Z))_\MULN(R_T_T_(I|Z|I|Z))_\MULN(R_T_T_(Z|I|I|Z))_\MULN(R_T_T_(I|Z|Z|I))_\MULN(R_T_T_(Z|Z|I|Z))_\MULN(R_T_T_(I|Z|Z|Z))_\MULN(R_T_T_(Z|I|Z|I))_\MULN(R_T_T_(Z|Z|Z|I))_\MULN(R_T_T_(Z|I|Z|Z))_\MULN(R_T_T_(Z|Z|Z|Z))] 1 1 9.3261E-01 2 1.2118E-02 3 2.2991E-02 4 3.1705E-03 5 1.1969E-02 6 2.7982E-03 7 1.0523E-02 8 1.3539E-03 9 1.7929E-04 10 3.4496E-04 11 3.3103E-04 12 6.9617E-04 13 1.2454E-04 14 1.2322E-04 15 1.3408E-04 16 2.3072E-04 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 0.0000 0.0000 1.0000 1.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 2.4526 2.0970 1.9748 1.7936 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1228.8353 Where the fit function is -2 * Log-likelihood of raw ordinal GROUP NUMBER: 4 Correlated liability model: inefficient approach; clinical group data MATRIX A This is a FULL matrix of order 1 by 4 1 2 3 4 1 15.0800 15.0800 18.1900 18.1900 MATRIX B This is a FULL matrix of order 1 by 1 1 1 1.0000 MATRIX C This is a FULL matrix of order 1 by 1 1 1 0.1701 MATRIX D This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.3432 -0.1997 -0.3432 -0.1997 MATRIX G This is a computed FULL matrix of order 1 by 1 [=B+C] 1 1 1.1701 MATRIX I This is an IDENTITY matrix of order 1 by 1 MATRIX J This is a FULL matrix of order 4 by 1 1 1 1.0000 2 1.0000 3 1.0000 4 1.0000 MATRIX K This is a computed FULL matrix of order 12 by 1 [=B.(\MULN(R_T_T_(I|Z|I|I)))_C.(\MULN(R_T_T_(Z|I|I|I)))_G.(\MULN(R_T_T_(Z|Z|I|I)))_B.(\MULN(R_T_T_(I|Z|I|Z)))_C.(\MULN(R_T_T_(Z|I|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|I)))_G.(\MULN(R_T_T_(Z|Z|I|Z)))_B.(\MULN(R_T_T_(I|Z|Z|Z)))_C.(\MULN(R_T_T_(Z|I|Z|I)))_G.(\MULN(R_T_T_(Z|Z|Z|I)))_C.(\MULN(R_T_T_(Z|I|Z|Z)))_G.(\MULN(R_T_T_(Z|Z|Z|Z)))] 1 1 2.4333E-02 2 2.5594E-03 3 1.5162E-02 4 3.2282E-03 5 1.7372E-04 6 8.2443E-04 7 2.1264E-03 8 1.8257E-03 9 1.1542E-04 10 7.6787E-04 11 1.3754E-04 12 1.5994E-03 MATRIX L This is a FULL matrix of order 1 by 4 1 2 3 4 1 4.1539 3.4084 4.1539 3.4084 MATRIX M This is a FULL matrix of order 1 by 1 1 1 0.0000 MATRIX N This is a computed FULL matrix of order 12 by 1 [=(K%U)] 1 1 0.4604 2 0.0484 3 0.2869 4 0.0611 5 0.0033 6 0.0156 7 0.0402 8 0.0345 9 0.0022 10 0.0145 11 0.0026 12 0.0303 MATRIX O This is a FULL matrix of order 1 by 4 1 2 3 4 1 -0.1320 -0.1017 -0.1320 -0.1017 MATRIX R This is a constrained FULL matrix of order 4 by 4 1 2 3 4 1 1.0000 0.7509 0.2322 0.2397 2 0.7509 1.0000 0.2397 0.3192 3 0.2322 0.2397 1.0000 0.7509 4 0.2397 0.3192 0.7509 1.0000 MATRIX S This is a FULL matrix of order 1 by 4 1 2 3 4 1 1.0000 1.0000 0.0000 0.0000 MATRIX T This is a computed FULL matrix of order 1 by 4 [=L+(A.O)+(S.D)] 1 2 3 4 1 1.8203 1.6745 1.7530 1.5578 MATRIX U This is a computed FULL matrix of order 12 by 1 [=Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y_Y] 1 1 0.0529 2 0.0529 3 0.0529 4 0.0529 5 0.0529 6 0.0529 7 0.0529 8 0.0529 9 0.0529 10 0.0529 11 0.0529 12 0.0529 MATRIX V This is a FULL matrix of order 1 by 1 1 1 0.1592 MATRIX Y This is a computed FULL matrix of order 1 by 1 [=\SUM(K)] 1 1 0.0529 MATRIX Z This is a NULL matrix of order 1 by 1 Matrix of EXPECTED thresholds DUMMY Threshold 1 0.0000 (OBSERVED MATRIX is nonexistent for raw data) EXPECTED COVARIANCE MATRIX DUMMY DUMMY 0.1592 Function value of this group: 1157.2917 Where the fit function is -2 * Log-likelihood of raw ordinal *** WARNING! *** Minimization may not be successful. See above CODE GREEN - it probably was OK Your model has 10 estimated parameters and 675 Observed statistics -2 times log-likelihood of data >>> 2386.127 Degrees of freedom >>>>>>>>>>>>>>>> 665 This problem used 0.8% of my workspace Task Time elapsed (DD:HH:MM:SS) Reading script & data 0: 0: 0: 0.11 Execution 0: 0: 8: 0.71 TOTAL 0: 0: 8: 0.82 Total number of warnings issued: 10 ______________________________________________________________________________ !save corrace.mxs ! no correlation model !drop 2 4 !end ______________________________________________________________________________