********************************************************************* * Replication file 10/11 for "Are Sports Teams Multiproduct Firms?" * by K.G. Stewart and J.C.H. Jones (2010) Empirical Economics 39(2), * 487-514. * Estimation of MSGM factor demand systems for baseball player * characteristics, the model M0. We estimate the nonseparable model * with concavity imposed. * This printout compute marginal costs and use them to obtain the * product-specific returns to scale values reported in Table 8, * including their standard errors. ********************************************************************* * Sample consists of 26 teams x 6 seasons (1986-91) SAMPLE 1 156; * Read in team characteristics; documented in Team.doc read(Team.dat) season team league attend attlag ticprice & pmarkup gameswon gameslst wins salaries nplayers strkouts walks & slugavg nhitters npitchrs experhit experpit nstars roadatt & aleast alcentrl alwest nleast nlcentrl nlwest * Delete variables not used in this analysis delete season team league attlag ticprice pmarkup gameswon gameslst salaries nplayers * read in hedonic prices read(HedPrice1.dat) p1 p2 p3 p4 nobs * define team outputs * This scaling of road attendance yields fastest convergence GENR Y1=WINS GENR Y2=ROADatt/10000000 * conference dummies genr dum1=aleast genr dum2=alcentrl genr dum3=alwest genr dum4=nleast genr dum5=nlcentrl * define team factor inputs: experience, hitting, pitching, stars GENR q1=experhit+experpit GENR q2=nhitters*slugavg GENR q3=npitchrs*strkouts/walks GENR q4=nstars * DEFLATE HEDONIC PRICES TO 1991 DOLLARS SAMPLE 1 26 GENR p1=p1*135.0/109.6 GENR p2=p2*135.0/109.6 GENR p3=p3*135.0/109.6 GENR p4=p4*135.0/109.6 SAMPLE 27 52 GENR p1=p1*135.0/113.6 GENR p2=p2*135.0/113.6 GENR p3=p3*135.0/113.6 GENR p4=p4*135.0/113.6 SAMPLE 53 78 GENR p1=p1*135.0/118.3 GENR p2=p2*135.0/118.3 GENR p3=p3*135.0/118.3 GENR p4=p4*135.0/118.3 SAMPLE 79 104 GENR p1=p1*135.0/124.0 GENR p2=p2*135.0/124.0 GENR p3=p3*135.0/124.0 GENR p4=p4*135.0/124.0 SAMPLE 105 130 GENR p1=p1*135.0/130.7 GENR p2=p2*135.0/130.7 GENR p3=p3*135.0/130.7 GENR p4=p4*135.0/130.7 SAMPLE 1 156 * Generate cost identity and cost shares * (This synthetic (hedonic) cost series is used * only for descriptive purposes in calculating the implied cost shares, not * in estimation of cost function parameters. It can be compared with fitted * cost series predicted by estimated model.) genr cost=q1*p1+q2*p2+q3*p3+q4*p4 genr s1=q1*p1/cost genr s2=q2*p2/cost genr s3=q3*p3/cost genr s4=q4*p4/cost stat Y1 Y2 p1 p2 p3 p4 s1 s2 s3 s4 * Obtain mean factor inputs: * qm1=135.73077 * qm2=5.35117 * qm3=17.92278 * qm4=2.61538 stat q1 / mean=qm1 stat q2 / mean=qm2 stat q3 / mean=qm3 stat q4 / mean=qm4 GENR DENOM=qm1*p1+qm2*p2+qm3*p3+qm4*p4 gen1 theta1=qm1/2 gen1 theta2=qm2/2 gen1 theta3=qm3/2 gen1 theta4=qm4/2 * generate price constructs used in estimation GENR Pd1=p1/DENOM GENR Pd2=p2/DENOM GENR Pd3=p3/DENOM GENR Pd4=p4/DENOM GENR Pd11=pd1*pd1 GENR Pd12=pd1*pd2 GENR Pd13=pd1*pd3 GENR Pd14=pd1*pd4 GENR Pd21=pd2*pd1 GENR Pd22=pd2*pd2 GENR Pd23=pd2*pd3 GENR Pd24=pd2*pd4 GENR Pd31=pd3*pd1 GENR Pd32=pd3*pd2 GENR Pd33=pd3*pd3 GENR Pd34=pd3*pd4 GENR Pd41=pd4*pd1 GENR Pd42=pd4*pd2 GENR Pd43=pd4*pd3 GENR Pd44=pd4*pd4 GENR Pd11=Pd11-Pd14-(Pd41-Pd44) GENR Pd12=Pd12-Pd14-(Pd42-Pd44) GENR Pd13=Pd13-Pd14-(Pd43-Pd44) GENR Pd21=Pd21-Pd24-(Pd41-Pd44) GENR Pd22=Pd22-Pd24-(Pd42-Pd44) GENR Pd23=Pd23-Pd24-(Pd43-Pd44) GENR Pd31=Pd31-Pd34-(Pd41-Pd44) GENR Pd32=Pd32-Pd34-(Pd42-Pd44) GENR Pd33=Pd33-Pd34-(Pd43-Pd44) GENR Pd12=2*Pd12 GENR Pd13=2*Pd13 GENR Pd23=2*Pd23 GENR Pd1=Pd1-Pd4 GENR Pd2=Pd2-Pd4 GENR Pd3=Pd3-Pd4 * Obtain parallel scalar means used in reconstruction of * total cost predicted at pt of variable means stat p1 / mean=m1 stat p2 / mean=m2 stat p3 / mean=m3 stat p4 / mean=m4 GEN1 m11=m1*m1 GEN1 m12=m1*m2 GEN1 m13=m1*m3 GEN1 m14=m1*m4 GEN1 m21=m2*m1 GEN1 m22=m2*m2 GEN1 m23=m2*m3 GEN1 m24=m2*m4 GEN1 m31=m3*m1 GEN1 m32=m3*m2 GEN1 m33=m3*m3 GEN1 m34=m3*m4 GEN1 m41=m4*m1 GEN1 m42=m4*m2 GEN1 m43=m4*m3 GEN1 m44=m4*m4 GEN1 m11=m11-m14-(m41-m44) GEN1 m12=m12-m14-(m42-m44) GEN1 m13=m13-m14-(m43-m44) GEN1 m21=m21-m24-(m41-m44) GEN1 m22=m22-m24-(m42-m44) GEN1 m23=m23-m24-(m43-m44) GEN1 m31=m31-m34-(m41-m44) GEN1 m32=m32-m34-(m42-m44) GEN1 m33=m33-m34-(m43-m44) * check symmetry: pairs should be equal print m12 m21 print m13 m31 print m23 m32 GEN1 m12=2*m12 GEN1 m13=2*m13 GEN1 m23=2*m23 GEN1 mDENOM=qm1*m1+qm2*m2+qm3*m3+qm4*m4 GEN1 mdenom2=mdenom*mdenom GEN1 md1=m1/mDENOM GEN1 md2=m2/mDENOM GEN1 md3=m3/mDENOM GEN1 md4=m4/mDENOM * Obtain parallel scalar means used in reconstruction of * factor demands and shares predicted at pt of variable means GEN1 md11=md1*md1 GEN1 md12=md1*md2 GEN1 md13=md1*md3 GEN1 md14=md1*md4 GEN1 md21=md2*md1 GEN1 md22=md2*md2 GEN1 md23=md2*md3 GEN1 md24=md2*md4 GEN1 md31=md3*md1 GEN1 md32=md3*md2 GEN1 md33=md3*md3 GEN1 md34=md3*md4 GEN1 md41=md4*md1 GEN1 md42=md4*md2 GEN1 md43=md4*md3 GEN1 md44=md4*md4 GEN1 md11=md11-md14-(md41-md44) GEN1 md12=md12-md14-(md42-md44) GEN1 md13=md13-md14-(md43-md44) GEN1 md21=md21-md24-(md41-md44) GEN1 md22=md22-md24-(md42-md44) GEN1 md23=md23-md24-(md43-md44) GEN1 md31=md31-md34-(md41-md44) GEN1 md32=md32-md34-(md42-md44) GEN1 md33=md33-md34-(md43-md44) GEN1 md12=2*md12 GEN1 md13=2*md13 GEN1 md23=2*md23 gen1 md1=md1-md4 gen1 md2=md2-md4 gen1 md3=md3-md4 stat y1 / mean=ym1 stat y2 / mean=ym2 ** INITIAL PARAMETERIZATION ** * Shazam requires that the theta's (/2) appear numerically * Multiplicative factor 2 is included in parameter D1 nl 4 / ncoef=41 iter=999 Eq q1=(S11*Pd1+S12*Pd2+S13*Pd3 & -67.8654*(S11*Pd11+S12*Pd12+S13*Pd13 & +S22*Pd22+S23*Pd23 & +S33*Pd33))*(Y1+BETA*Y2) & +d11*dum1+d12*dum2+d13*dum3+d14*dum4+d15*dum5 & +B11*(Y1+BETA*Y2)+B1+LAMBDA1*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q2=(S12*Pd1+S22*Pd2+S23*Pd3 & -2.67558*(S11*Pd11+S12*Pd12+S13*Pd13 & +S22*Pd22+S23*Pd23 & +S33*Pd33))*(Y1+BETA*Y2) & +d21*dum1+d22*dum2+d23*dum3+d24*dum4+d25*dum5 & +B22*(Y1+BETA*Y2)+B2+LAMBDA2*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q3=(S13*Pd1+S23*Pd2+S33*Pd3 & -8.96111*(S11*Pd11+S12*Pd12+S13*Pd13 & +S22*Pd22+S23*Pd23 & +S33*Pd33))*(Y1+BETA*Y2) & +d31*dum1+d32*dum2+d33*dum3+d34*dum4+d35*dum5 & +B33*(Y1+BETA*Y2)+B3+LAMBDA3*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q4=-((S11+S12+S13)*Pd1+(S12+S22+S23)*Pd2+(S13+S23+S33)*Pd3 & +1.30769*(S11*Pd11+S12*Pd12+S13*Pd13 & +S22*Pd22+S23*Pd23 & +S33*Pd33))*(Y1+BETA*Y2) & +d41*dum1+d42*dum2+d43*dum3+d44*dum4+d45*dum5 & +B44*(Y1+BETA*Y2)+B4+LAMBDA4*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) end * Check integrity of components of the cost expression that will be used * in elasticities by computing: * Cost function at pt of variable means test ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom) & +b11*m1+b22*m2+b33*m3+b44*m4)*(ym1+beta*ym2)+b1*m1+b2*m2+b3*m3+b4*m4 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2+D2*Ym2*Ym2) * pscale1=1-((C(y1,y2)-C(0,y2))/(MC1*y1)) test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*(ym1+beta*ym2) & +(b1*m1+b2*m2+b3*m3+b4*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2+D2*Ym2*Ym2) & -((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*(beta*ym2) & -(b1*m1+b2*m2+b3*m3+b4*m4) & -(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D2*Ym2*Ym2) )*ym1/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(2*Ym1+D1*Ym2) ) * pscale1: sensitivity to reparameterization test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*ym1 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2) )*ym1/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(2*Ym1+D1*Ym2) ) * pscale2=1-((C(y1,y2)-C(y1,0))/(MC2*y2)) test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*(ym1+beta*ym2) & +(b1*m1+b2*m2+b3*m3+b4*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2+D2*Ym2*Ym2) & -((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*ym1 & -(b1*m1+b2*m2+b3*m3+b4*m4) & -(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1) )*ym2/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1+2*D2*Ym2) ) * pscale2: sensitivity to reparameterization test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta*ym2 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1*Ym2+D2*Ym2*Ym2) )*ym2/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1+2*D2*Ym2) ) ** ALTERNATIVE (BUT EQUIVALENT) PARAMETERIZATION ** GENR Z11=Pd1-THETA1*Pd11 GENR Z12=Pd2-THETA1*Pd12 GENR Z13=Pd3-THETA1*Pd13 GENR Z14=-THETA1*Pd22 GENR Z15=-THETA1*Pd23 GENR Z16=-THETA1*Pd33 GENR Z21=-THETA2*Pd11 GENR Z22=Pd1-THETA2*Pd12 GENR Z23=-THETA2*Pd13 GENR Z24=Pd2-THETA2*Pd22 GENR Z25=Pd3-THETA2*Pd23 GENR Z26=-THETA2*Pd33 GENR Z31=-THETA3*Pd11 GENR Z32=-THETA3*Pd12 GENR Z33=Pd1-THETA3*Pd13 GENR Z34=-THETA3*Pd22 GENR Z35=Pd2-THETA3*Pd23 GENR Z36=Pd3-THETA3*Pd33 GENR Z41=-(Pd1+THETA4*Pd11) GENR Z42=-(Pd1+Pd2+THETA4*Pd12) GENR Z43=-(Pd1+Pd3+THETA4*Pd13) GENR Z44=-(Pd2+THETA4*Pd22) GENR Z45=-(Pd2+Pd3+THETA4*Pd23) GENR Z46=-(Pd3+THETA4*Pd33) * Multiplicative factor 2 is included in parameter D1 nl 4 / ncoef=41 iter=999 Eq q1=(Y1+BETA*Y2) & *(B11+S11*Z11+S12*Z12+S13*Z13+S22*Z14+S23*Z15+S33*Z16) & +d11*dum1+d12*dum2+d13*dum3+d14*dum4+d15*dum5 & +B1+LAMBDA1*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q2=(Y1+BETA*Y2) & *(B22+S11*Z21+S12*Z22+S13*Z23+S22*Z24+S23*Z25+S33*Z26) & +d21*dum1+d22*dum2+d23*dum3+d24*dum4+d25*dum5 & +B2+LAMBDA2*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q3=(Y1+BETA*Y2) & *(B33+S11*Z31+S12*Z32+S13*Z33+S22*Z34+S23*Z35+S33*Z36) & +d31*dum1+d32*dum2+d33*dum3+d34*dum4+d35*dum5 & +B3+LAMBDA3*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q4=(Y1+BETA*Y2) & *(B44+S11*Z41+S12*Z42+S13*Z43+S22*Z44+S23*Z45+S33*Z46) & +d41*dum1+d42*dum2+d43*dum3+d44*dum4+d45*dum5 & +B4+LAMBDA4*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) end * Check integrity of components of the cost expression that will be used * in elasticities by computing: * Cost function at pt of variable means test ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom) & +b11*m1+b22*m2+b33*m3+b44*m4)*(ym1+beta*ym2)+b1*m1+b2*m2+b3*m3+b4*m4 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2+D2*Ym2*Ym2) * pscale1=1-((C(y1,y2)-C(0,y2))/(MC1*y1)) test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*(ym1+beta*ym2) & +(b1*m1+b2*m2+b3*m3+b4*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2+D2*Ym2*Ym2) & -((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*(beta*ym2) & -(b1*m1+b2*m2+b3*m3+b4*m4) & -(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D2*Ym2*Ym2) )*ym1/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(2*Ym1+D1*Ym2) ) * pscale1: sensitivity to reparameterization test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*ym1 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2) )*ym1/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(2*Ym1+D1*Ym2) ) * pscale2=1-((C(y1,y2)-C(y1,0))/(MC2*y2)) test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*(ym1+beta*ym2) & +(b1*m1+b2*m2+b3*m3+b4*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2+D2*Ym2*Ym2) & -((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*ym1 & -(b1*m1+b2*m2+b3*m3+b4*m4) & -(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1) )*ym2/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1+2*D2*Ym2) ) * pscale2: sensitivity to reparameterization test 1- ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta*ym2 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1*Ym2+D2*Ym2*Ym2) )*ym2/ & ( ((S11*m11+S12*m12+S13*m13 & +S22*m22+S23*m23 & +S33*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1+2*D2*Ym2) ) ** CONCAVITY IMPOSED USING WILEY-SCHMIDT-BRAMBLE REPARAMETERIZATION ** * Multiplicative factor 2 is included in parameter D1: this turns out to be * critical to the quick convergence of the concavity-constrained system nl 4 / ncoef=41 iter=999 coef=g predict=qpred Eq q1=-(G11*G11*Z11+G11*G21*Z12+G11*G31*Z13+(G21*G21+G22*G22)*Z14 & +(G31*G21+G32*G22)*Z15+(G31*G31+G32*G32+G33*G33)*Z16-B11)*(Y1+BETA*Y2) & +d11*dum1+d12*dum2+d13*dum3+d14*dum4+d15*dum5 & +B1+LAMBDA1*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q2=-(G11*G11*Z21+G11*G21*Z22+G11*G31*Z23+(G21*G21+G22*G22)*Z24 & +(G31*G21+G32*G22)*Z25+(G31*G31+G32*G32+G33*G33)*Z26-B22)*(Y1+BETA*Y2) & +d21*dum1+d22*dum2+d23*dum3+d24*dum4+d25*dum5 & +B2+LAMBDA2*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q3=-(G11*G11*Z31+G11*G21*Z32+G11*G31*Z33+(G21*G21+G22*G22)*Z34 & +(G31*G21+G32*G22)*Z35+(G31*G31+G32*G32+G33*G33)*Z36-B33)*(Y1+BETA*Y2) & +d31*dum1+d32*dum2+d33*dum3+d34*dum4+d35*dum5 & +B3+LAMBDA3*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) Eq q4=-(G11*G11*Z41+G11*G21*Z42+G11*G31*Z43+(G21*G21+G22*G22)*Z44 & +(G31*G21+G32*G22)*Z45+(G31*G31+G32*G32+G33*G33)*Z46-B44)*(Y1+BETA*Y2) & +d41*dum1+d42*dum2+d43*dum3+d44*dum4+d45*dum5 & +B4+LAMBDA4*(Y1*Y1+D1*Y1*Y2+D2*Y2*Y2) end * We use the following parameter equivalencies * s11=-g11*g11 * s12=-g11*g21 * s13=-g11*g31 * s22=-(g21*g21+g22*g22) * s23=-(g21*g31+g22*g32) * s33=-(g31*g31+g32*g32+g33*g33) * pscale1 (reparameterized) test 1- ( ((-g11*g11*m11-g11*g21*m12-g11*g31*m13 & -(g21*g21+g22*g22)*m22-(g21*g31+g22*g32)*m23 & -(g31*g31+g32*g32+g33*g33)*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*ym1 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(Ym1*Ym1+D1*Ym1*Ym2) )*ym1/ & ( ((-g11*g11*m11-g11*g21*m12-g11*g31*m13 & -(g21*g21+g22*g22)*m22-(g21*g31+g22*g32)*m23 & -(g31*g31+g32*g32+g33*g33)*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4) & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(2*Ym1+D1*Ym2) ) * pscale2 (reparameterized) test 1- ( ((-g11*g11*m11-g11*g21*m12-g11*g31*m13 & -(g21*g21+g22*g22)*m22-(g21*g31+g22*g32)*m23 & -(g31*g31+g32*g32+g33*g33)*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta*ym2 & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1*Ym2+D2*Ym2*Ym2) )*ym2/ & ( ((-g11*g11*m11-g11*g21*m12-g11*g31*m13 & -(g21*g21+g22*g22)*m22-(g21*g31+g22*g32)*m23 & -(g31*g31+g32*g32+g33*g33)*m33)/(2*mdenom)+b11*m1+b22*m2+b33*m3+b44*m4)*beta & +(lambda1*m1+lambda2*m2+lambda3*m3+lambda4*m4)*(D1*Ym1+2*D2*Ym2) ) stop