*--------------------------------------------------------------* * Test Problem 5 from Chapter 8, section 5.7 * * Kowalik Problem * *--------------------------------------------------------------* Sets m number of data sets /1*11/ n number of variabels /1*2/ p number of parameters /1*4/; Parameters ze(m,n) observed data values z2(m) model constant std(n) standard deviations; Variables z(m) fitted data variables a(p) model parameters c cost function; table ze(m,n) 1 2 1 0.1957 0.25 2 0.1947 0.5 3 0.1735 1.0 4 0.1600 2.0 5 0.0844 4.0 6 0.0627 6.0 7 0.0456 8.0 8 0.0342 10.0 9 0.0323 12.0 10 0.0235 14.0 11 0.0246 16.0; z2(m) = 1/ze(m,'2'); std(n) = 1; Equations obj objective function con(m) model constraint; obj.. c =e= sum(m,sqr((z(m)-ze(m,'1'))/std('1'))); con(m) .. -z(m) + a('1')*(sqr(z2(m)) + z2(m)*a('2')) /(sqr(z2(m)) + z2(m)*a('3') + a('4')) =e= 0; model problem /obj,con/; z.lo(m) = ze(m,'1') - 0.02; z.up(m) = ze(m,'1') + 0.02; a.lo(p) = -0.2892; a.up(p) = 0.2893; z.l(m) = uniform(z.lo(m),z.up(m)); a.l(p) = uniform(a.lo(p), a.up(p)); solve problem using nlp minimizing c;