Re: Problem with NMaximize

*To*: mathgroup at smc.vnet.net*Subject*: [mg90044] Re: Problem with NMaximize*From*: DrMajorBob <drmajorbob at att.net>*Date*: Fri, 27 Jun 2008 06:16:25 -0400 (EDT)*References*: <g3vkvi$kio$1@smc.vnet.net> <4863CA88.5040100@gmail.com>*Reply-to*: drmajorbob at longhorns.com

NSolve and NMaximize generally use a kind of secant search, using previous attempts to extrapolate to a new and (hopefully) better one. Unfortunately, extrapolation can easily carry us from feasible points to infeasible ones. If you can formulate the problem so that infeasaible points can't exist (or can't be reached by extrapolation), you avoid a lot of trouble. Anyway, why maximize log-likelihood, when you could maximize the likelihood itself? If THAT turns out to be negative at an intermediate step, it's not a problem, since you're not taking a logarithm of it. We generally speak of log-likelihood because of its use in an asymptotic chi-square test, but that's not always relevant when choosing optimum parameters. Bobby On Thu, 26 Jun 2008 13:45:57 -0500, ramiro <ramiro.barrantes at uvm.edu> wrote: > Thank you very much for your responses, Bobby and Jean-Marc. I did try > changing the parametrization (thanks Bobby!) and I got an answer. > However, let me try again with a hopefully more clear example, as I > don't understand why this doesn't work, in particular, I don't > understand why is it evaluating negative numbers if I am putting as a > constraint that it shouldn't? > > > Sample likelihood, > > In[196]:= > logLikelihood1[t1_?NumberQ, t2_?NumberQ, t3_?NumberQ, t4_?NumberQ, > t5_?NumberQ, t6_?NumberQ, t7_?NumberQ, t8_?NumberQ, t9_?NumberQ, > t10_?NumberQ, t11_?NumberQ, t12_?NumberQ] := > Log[0.00001712341312000713` t1 + 5.648647024829866`*^-6 t10 + > 8.90597537441381`*^-6 t11 + 1.9727284288726167`*^-6 t12 + > 4.102725237468317`*^-6 t2 + 3.7864902508468615`*^-6 t3 + > 9.215772325326653`*^-7 t4 + 0.000057161484895917856` t5 + > 0.000012892953334779516` t6 + 8.646320198720343`*^-6 t7 + > 5.877910295858781`*^-6 t8 + 1.6837835562631724`*^-6 t9]; > > Another version (without the NumberQ)... > > In[197]:= > logLikelihood2[t1_, t2_, t3_, t4_, t5_, t6_, t7_, t8_, t9_, t10_, > t11_, t12_] := > Log[0.00001712341312000713` t1 + 5.648647024829866`*^-6 t10 + > 8.90597537441381`*^-6 t11 + 1.9727284288726167`*^-6 t12 + > 4.102725237468317`*^-6 t2 + 3.7864902508468615`*^-6 t3 + > 9.215772325326653`*^-7 t4 + 0.000057161484895917856` t5 + > 0.000012892953334779516` t6 + 8.646320198720343`*^-6 t7 + > 5.877910295858781`*^-6 t8 + 1.6837835562631724`*^-6 t9]; > > In[198]:= NMaximize[{logLikelihood1[t1, t2, t3, t4, t5, t6, t7, t8, > t9, t10, t11, t12], t1 > 0, t2 > 0, t3 > 0, t4 > 0, t5 > 0, t6 > 0, > t7 > 0, t8 > 0, t9 > 0, t10 > 0, t11 > 0, t12 > 0, > t1 + t10 + t11 + t12 + t2 + t3 + t4 + t5 + t6 + t7 + t8 + t9 == > 1}, {t1, t2, t3, t4, t5, t6, t7, t8, t9, t10, t11, t12}] > > During evaluation of In[198]:= NMaximize::nrnum: The function value \ > 15.7729-3.14159 I is not a real number at \ > {t1,t10,t11,t12,t2,t3,t4,t5,t6,t7,<<2>>} = {-0.490249,<<9>>,<<2>>}. \ > >> > > Out[198]= {-13.2945, {t1 -> 2.10942*10^-15, t10 -> -2.22045*10^-16, > t11 -> -2.22045*10^-16, t12 -> -2.22045*10^-16, > t2 -> -1.11022*10^-16, t3 -> -2.22045*10^-16, t4 -> -2.22045*10^-16, > t5 -> -6.93889*10^-18, t6 -> -2.22045*10^-16, > t7 -> -1.11022*10^-16, t8 -> -4.44089*10^-16, t9 -> 1.}} > > In[199]:= NMaximize[{logLikelihood2[t1, t2, t3, t4, t5, t6, t7, t8, > t9, t10, t11, t12], t1 > 0, t2 > 0, t3 > 0, t4 > 0, t5 > 0, t6 > 0, > t7 > 0, t8 > 0, t9 > 0, t10 > 0, t11 > 0, t12 > 0, > t1 + t2 + t3 + t4 + t5 + t6 + t7 + t8 + t9 + t10 + t11 + t12 == > 1}, {t1, t2, t3, t4, t5, t6, t7, t8, t9, t10, t11, t12}] > > During evaluation of In[199]:= NMaximize::nrnum: The function value \ > 15.7729-3.14159 I is not a real number at \ > {t1,t10,t11,t12,t2,t3,t4,t5,t6,t7,<<2>>} = {-0.490249,<<9>>,<<2>>}. \ > >> > > Out[199]= {-13.2945, {t1 -> 2.10942*10^-15, t10 -> -2.22045*10^-16, > t11 -> -2.22045*10^-16, t12 -> -2.22045*10^-16, > t2 -> -1.11022*10^-16, t3 -> -2.22045*10^-16, t4 -> -2.22045*10^-16, > t5 -> -6.93889*10^-18, t6 -> -2.22045*10^-16, > t7 -> -1.11022*10^-16, t8 -> -4.44089*10^-16, t9 -> 1.}} > -- DrMajorBob at longhorns.com