From rickyyao at gmail.com Fri May 1 07:21:36 2026 From: rickyyao at gmail.com (Richard Yao) Date: Fri, 1 May 2026 09:21:36 +1200 Subject: [Limdep Nlogit List] Modelling attribute non-attendance (ANA) in NLOGIT 6 In-Reply-To: References: Message-ID: Dear members, I hope you are all doing well. I am writing to seek your advice on modelling attribute non-attendance (ANA) in NLOGIT 6. I have included my code and output below. While the code appears to run correctly, the results seem unusual?the estimated coefficients are identical across all four latent classes. I have previously implemented ANA models for an ERE paper using an earlier version of NLOGIT, where the results behaved as expected. I would greatly appreciate any guidance you may be able to provide. Many thanks in advance. Best regards, Richard N = 509 |-> LOAD;file="C:\Users\yaor\CE_LCM_Data_8Apr2026_n_509.lpj"$ Project file contained 8958 observations. |-> NLOGIT ;lhs = chosen,cset,altij ;choices=sq,a,b ;maxit=200 ? panel's length & Max. num. of iterations ;lcm = a_sbr_1, a_uni_hi, a_100abo, a1_comn, a3_drun, a7_worn, p1_lat, s4_ind, i3_bre ;pts = 4 ? specifies it is a LCM and the number of classes ;pds=times ? specifies the log-likelihood with the product operator,taking the product of the probabilities from 6 choice sets ;rst= beh1,beh2,bmn1,bmn2,bpol1,bpol2,bcost, baltd, ? Full attendance beh1,beh2,0, 0, bpol1,bpol2,0 , 0 , ? Ignoring boat efficiency, cost, & baltd 0 ,0 ,bmn1,0 ,0 ,0 ,bcost, baltd, ? Ignoring beh1, beh2, bmn2, bpol1, bpol2 0 ,0 ,0 ,0 ,0 ,0 ,0 , 0 ? Ignoring all attributes ;parameters ? estimate individual specific parameters ; Check Data ;model: U(sq)= beh1*e_h1 + beh2*e_h2 + bmn1*mn1 + bmn2*mn2 + bpol1*pol1 + bpol2*pol2 + bcost*cost / U(a) = b_altd*alt_d + beh1*e_h1 + beh2*e_h2 + bmn1*mn1 + bmn2*mn2 + bpol1*pol1 + bpol2*pol2 + bcost*cost / U(b) = b_altd*alt_d + beh1*e_h1 + beh2*e_h2 + bmn1*mn1 + bmn2*mn2 + bpol1*pol1 + bpol2*pol2 + bcost*cost $ +----------------------------------------------------------+ | Inspecting the data set before estimation. | | These errors mark observations which will be skipped. | | Row Individual = 1st row then group number of data block | +----------------------------------------------------------+ No bad observations were found in the sample Iterative procedure has converged Normal exit: 6 iterations. Status=0, F= .2438144D+04 ----------------------------------------------------------------------------- Discrete choice (multinomial logit) model Dependent variable Choice Log likelihood function -2438.14357 Estimation based on N = 2986, K = 8 Inf.Cr.AIC = 4892.3 AIC/N = 1.638 --------------------------------------- Log likelihood R-sqrd R2Adj ASCs only model must be fit separately Use NLOGIT ;...;RHS=ONE$ Note: R-sqrd = 1 - logL/Logl(constants) Warning: Model does not contain a full set of ASCs. R-sqrd is problematic. Use model setup with ;RHS=one to get LogL0. Root Likelihood:Geom. Mean of P^ .4420 --------------------------------------- Response data are given as ind. choices Number of obs.= 2986, skipped 0 obs --------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence CHOSEN| Coefficient Error z |z|>Z* Interval --------+-------------------------------------------------------------------- BEH1|1| 1.10202*** .08669 12.71 .0000 .93211 1.27193 BEH2|1| 1.30316*** .06514 20.01 .0000 1.17549 1.43082 BMN1|1| .79582*** .08215 9.69 .0000 .63481 .95682 BMN2|1| .35273*** .06870 5.13 .0000 .21808 .48737 BPOL1|1| .61010*** .09302 6.56 .0000 .42779 .79241 BPOL2|1| .47487*** .06612 7.18 .0000 .34528 .60445 BCOST|1| -.00079*** .00012 -6.43 .0000 -.00103 -.00055 B_ALTD|1| .05026 .10817 .46 .6422 -.16176 .26228 --------+-------------------------------------------------------------------- ***, **, * ==> Significance at 1%, 5%, 10% level. Model was estimated on Apr 10, 2026 at 05:47:01 PM ----------------------------------------------------------------------------- Iterative procedure has converged Normal exit: 49 iterations. Status=0, F= .2121261D+04 ----------------------------------------------------------------------------- Latent Class Logit Model Dependent variable CHOSEN Log likelihood function -2121.26079 Restricted log likelihood -3280.45629 Chi squared [ 38](P= .000) 2318.39102 Significance level .00000 McFadden Pseudo R-squared .3533641 Estimation based on N = 2986, K = 38 Inf.Cr.AIC = 4318.5 AIC/N = 1.446 --------------------------------------- Log likelihood R-sqrd R2Adj No coefficients -3280.4563 .3534 .3492 Constants only can be computed directly Use NLOGIT ;...;RHS=ONE$ At start values -2460.3355 .1378 .1323 Note: R-sqrd = 1 - logL/Logl(constants) Root Likelihood:Geom. Mean of P^ .4914 Warning: Model does not contain a full set of ASCs. R-sqrd is problematic. Use model setup with ;RHS=one to get LogL0. --------------------------------------- Response data are given as ind. choices Number of latent classes = 4 Average Class Probabilities .282 .400 .150 .168 LCM model with panel has 509 groups Variable number of obs./group =TIMES Number of obs.= 2986, skipped 0 obs --------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence CHOSEN| Coefficient Error z |z|>Z* Interval --------+-------------------------------------------------------------------- |Random utility parameters in latent class -->> 1............... BEH1|1| 1.92247*** .13203 14.56 .0000 1.66369 2.18125 BEH2|1| 2.54316*** .11962 21.26 .0000 2.30872 2.77761 BMN1|1| 1.37009*** .19790 6.92 .0000 .98221 1.75797 BMN2|1| 1.52454*** .19107 7.98 .0000 1.15005 1.89903 BPOL1|1| 1.38487*** .15370 9.01 .0000 1.08363 1.68611 BPOL2|1| 1.15980*** .10383 11.17 .0000 .95629 1.36331 BCOST|1| -.00570*** .00074 -7.71 .0000 -.00715 -.00425 B_ALTD|1| 2.56508*** .41237 6.22 .0000 1.75686 3.37331 |Random utility parameters in latent class -->> 2............... BEH1|2| 1.92247*** .13203 14.56 .0000 1.66369 2.18125 BEH2|2| 2.54316*** .11962 21.26 .0000 2.30872 2.77761 BMN1|2| 0.0 .....(Fixed Parameter)..... BMN2|2| 0.0 .....(Fixed Parameter)..... BPOL1|2| 1.38487*** .15370 9.01 .0000 1.08363 1.68611 BPOL2|2| 1.15980*** .10383 11.17 .0000 .95629 1.36331 BCOST|2| 0.0 .....(Fixed Parameter)..... B_ALTD|2| 0.0 .....(Fixed Parameter)..... |Random utility parameters in latent class -->> 3............... BEH1|3| 0.0 .....(Fixed Parameter)..... BEH2|3| 0.0 .....(Fixed Parameter)..... BMN1|3| 1.37009*** .19790 6.92 .0000 .98221 1.75797 BMN2|3| 0.0 .....(Fixed Parameter)..... BPOL1|3| 0.0 .....(Fixed Parameter)..... BPOL2|3| 0.0 .....(Fixed Parameter)..... BCOST|3| -.00570*** .00074 -7.71 .0000 -.00715 -.00425 B_ALTD|3| 2.56508*** .41237 6.22 .0000 1.75686 3.37331 |Random utility parameters in latent class -->> 4............... BEH1|4| 0.0 .....(Fixed Parameter)..... BEH2|4| 0.0 .....(Fixed Parameter)..... BMN1|4| 0.0 .....(Fixed Parameter)..... BMN2|4| 0.0 .....(Fixed Parameter)..... BPOL1|4| 0.0 .....(Fixed Parameter)..... BPOL2|4| 0.0 .....(Fixed Parameter)..... BCOST|4| 0.0 .....(Fixed Parameter)..... B_ALTD|4| 0.0 .....(Fixed Parameter)..... |This is THETA(01) in class probability model.................... _ONE|1| -1.46308 1.59698 -.92 .3596 -4.59310 1.66694 _A_SBR|1| .33726 .41054 .82 .4114 -.46739 1.14191 _A_UNI|1| .30410 .38328 .79 .4275 -.44713 1.05532 _A_100|1| .57732 .38364 1.50 .1324 -.17459 1.32924 _A1_CO|1| 1.16015 .98067 1.18 .2368 -.76193 3.08224 _A3_DR|1| -1.43785* .81865 -1.76 .0790 -3.04237 .16667 _A7_WO|1| .79912 1.25996 .63 .5259 -1.67036 3.26860 _P1_LA|1| -.72897 .92499 -.79 .4307 -2.54192 1.08399 _S4_IN|1| .86017 .89804 .96 .3382 -.89997 2.62030 _I3_BR|1| 1.07985 1.33400 .81 .4182 -1.53474 3.69445 |This is THETA(02) in class probability model.................... _ONE|2| -5.12052** 2.08257 -2.46 .0139 -9.20228 -1.03875 _A_SBR|2| .76384 .48705 1.57 .1168 -.19075 1.71844 _A_UNI|2| .60173 .44629 1.35 .1776 -.27297 1.47644 _A_100|2| .52510 .45145 1.16 .2448 -.35972 1.40992 _A1_CO|2| 1.28256 1.11903 1.15 .2517 -.91069 3.47581 _A3_DR|2| -2.90244*** 1.05502 -2.75 .0059 -4.97024 -.83464 _A7_WO|2| .96414 1.58739 .61 .5436 -2.14709 4.07537 _P1_LA|2| -2.39047** 1.01637 -2.35 .0187 -4.38252 -.39842 _S4_IN|2| 1.17789 1.05349 1.12 .2635 -.88690 3.24269 _I3_BR|2| 6.66912*** 1.86589 3.57 .0004 3.01204 10.32620 |This is THETA(03) in class probability model.................... _ONE|3| -1.69662 2.22571 -.76 .4459 -6.05893 2.66569 _A_SBR|3| 1.25179* .70877 1.77 .0774 -.13736 2.64095 _A_UNI|3| .39719 .49943 .80 .4264 -.58167 1.37604 _A_100|3| -.03146 .53379 -.06 .9530 -1.07768 1.01475 _A1_CO|3| -.24277 1.32231 -.18 .8543 -2.83445 2.34892 _A3_DR|3| -.89400 1.05243 -.85 .3956 -2.95673 1.16872 _A7_WO|3| 2.82413 1.85429 1.52 .1278 -.81021 6.45848 _P1_LA|3| -1.28405 1.20520 -1.07 .2867 -3.64619 1.07809 _S4_IN|3| -.14264 1.11107 -.13 .8978 -2.32030 2.03503 _I3_BR|3| -.06089 1.83862 -.03 .9736 -3.66452 3.54274 |This is THETA(04) in class probability model.................... _ONE|4| 0.0 .....(Fixed Parameter)..... _A_SBR|4| 0.0 .....(Fixed Parameter)..... _A_UNI|4| 0.0 .....(Fixed Parameter)..... _A_100|4| 0.0 .....(Fixed Parameter)..... _A1_CO|4| 0.0 .....(Fixed Parameter)..... _A3_DR|4| 0.0 .....(Fixed Parameter)..... _A7_WO|4| 0.0 .....(Fixed Parameter)..... _P1_LA|4| 0.0 .....(Fixed Parameter)..... _S4_IN|4| 0.0 .....(Fixed Parameter)..... _I3_BR|4| 0.0 .....(Fixed Parameter)..... --------+-------------------------------------------------------------------- ***, **, * ==> Significance at 1%, 5%, 10% level. Fixed parameter ... is constrained to equal the value or had a nonpositive st.error because of an earlier problem. Model was estimated on Apr 10, 2026 at 05:47:07 PM -----------------------------------------------------------------------------