[Limdep Nlogit List] Interpretation of interactions
Harold Mayaba
mayabach2009 at yahoo.com
Wed Nov 2 18:39:13 AEDT 2022
Dear Professor Greene and Nlogit users,
Thank you for the continued help. I have made some progress from the time I posted on the forum, see the results below. I have the following questions:
1. I have interacted sociodemographic characteristics with some selected variables, please help me understand how to interpret these results?
2. I would like to understand the probability of choosing a 'no buy option,' to achieve this I have included the effect commander, but I'm not sure if I have done the right thing. I have obtained these results below, please help me to interpret the results.
Elasticity wrt change of X in row choice on Prob[column choice]--------+--------------------------CAGED | ALTI1 ALTI2 NONE--------+-------------------------- ALTI1| .0784 -.0557 .0020 ALTI2| -.0522 .0635 .0042 NONE| .0000 .0000 .0000
3. Please help me check all the results and advise if there are some areas that needs to be worked on.
Regards
Harold
NLOGIT
; lhs = choice
; choices = alti1,alti2, none
;rpl = female,male, agea, ageb, agec, aged, agee
; fcn=caged(n) ,acdoors(n),nprs(n|#0000000),cert1(n|#0000000),cert2(n|#0000000) ,mmort(n|#0000000),hmort(n|#0000000)
; pds =10
; pts=100
;maxit=500
;describe
;parameters
; halton
;effects: caged (*) +acdoors(*)
;wtp=caged/price,acdoors/price ,nprs/price ,cert1/price,cert2/price
,mmort/price,hmort/price
; model:
U(alti1,alti2) = caged*caged+acdoors*acdoors+nprs*nprs+cert1*cert1+ cert2*cert2+mmort*mmort+hmort*hmort + price*price/
U(None) = ASC + caged*caged+acdoors*acdoors+nprs*nprs+cert1*cert1+cert2*cert2+mmort*mmort+hmort*hmort + price*price
-----------------------------------------------------------------------------Start values obtained using MNL modelDependent variable ChoiceLog likelihood function -8530.68923Estimation based on N = 9550, K = 9Inf.Cr.AIC = 17079.4 AIC/N = 1.788--------------------------------------- Log likelihood R-sqrd R2AdjConstants only ********** .1742 .1729Note: R-sqrd = 1 - logL/Logl(constants)Warning: Model does not contain a fullset of ASCs. R-sqrd is problematic. Usemodel setup with ;RHS=one to get LogL0.---------------------------------------Response data are given as ind. choicesNumber of obs.= 9550, skipped 0 obs--------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence CHOICE| Coefficient Error z |z|>Z* Interval--------+-------------------------------------------------------------------- CAGED| -.77011*** .06145 -12.53 .0000 -.89056 -.64967 ACDOORS| 1.45760*** .04171 34.95 .0000 1.37585 1.53935 NPRS| .18781*** .07284 2.58 .0099 .04505 .33057 CERT1| .10802 .06642 1.63 .1039 -.02215 .23819 CERT2| .13654** .05770 2.37 .0180 .02345 .24963 MMORT| -.25636*** .03965 -6.47 .0000 -.33406 -.17865 HMORT| -.73000*** .03872 -18.85 .0000 -.80590 -.65411 PRICE| -.11309*** .00602 -18.78 .0000 -.12489 -.10129 ASC| -.38268*** .11405 -3.36 .0008 -.60621 -.15914--------+--------------------------------------------------------------------***, **, * ==> Significance at 1%, 5%, 10% level.Model was estimated on Nov 02, 2022 at 07:38:37 PM-----------------------------------------------------------------------------
Iterative procedure has convergedNormal exit: 87 iterations. Status=0, F= .6891570D+04
-----------------------------------------------------------------------------Random Parameters Multinom. Logit ModelDependent variable CHOICELog likelihood function -6891.57039Restricted log likelihood -10491.74736Chi squared [ 30](P= .000) 7200.35393Significance level .00000McFadden Pseudo R-squared .3431437Estimation based on N = 9550, K = 30Inf.Cr.AIC = 13843.1 AIC/N = 1.450--------------------------------------- Log likelihood R-sqrd R2AdjNo coefficients ********** .3431 .3421Constants only ********** .3329 .3318At start values -8530.6892 .1921 .1909Note: R-sqrd = 1 - logL/Logl(constants)Warning: Model does not contain a fullset of ASCs. R-sqrd is problematic. Usemodel setup with ;RHS=one to get LogL0.---------------------------------------Response data are given as ind. choicesReplications for simulated probs. = 100Used Halton sequences in simulations.RPL model with panel has 955 groupsFixed number of obsrvs./group= 10Number of obs.= 9550, skipped 0 obs--------+-------------------------------------------------------------------- | Standard Prob. 95% Confidence CHOICE| Coefficient Error z |z|>Z* Interval--------+-------------------------------------------------------------------- |Random parameters in utility functions.............................. CAGED| -3.48684*** .53778 -6.48 .0000 -4.54088 -2.43281 ACDOORS| 2.32810*** .25424 9.16 .0000 1.82981 2.82640 NPRS| .05060 .12796 .40 .6925 -.20019 .30140 CERT1| .19253** .09454 2.04 .0417 .00723 .37782 CERT2| .22199*** .08397 2.64 .0082 .05741 .38657 MMORT| -.48682*** .05578 -8.73 .0000 -.59614 -.37750 HMORT| -1.31326*** .06501 -20.20 .0000 -1.44068 -1.18584 |Nonrandom parameters in utility functions........................... PRICE| -.19124*** .00860 -22.25 .0000 -.20809 -.17439 ASC| -1.25604*** .16647 -7.55 .0000 -1.58231 -.92978 |Heterogeneity in mean, Parameter:Variable...........................CAGE:FEM| .18451 .38670 .48 .6333 -.57341 .94244CAGE:MAL| .04152 .28127 .15 .8826 -.50976 .59281CAGE:AGE| .06301*** .01890 3.33 .0009 .02596 .10006CAG0:AGE| .02887** .01299 2.22 .0263 .00341 .05433CAG1:AGE| .00062 .00946 .07 .9478 -.01792 .01916CAG2:AGE| .01167 .00713 1.64 .1018 -.00231 .02565CAG3:AGE| .01258 .01011 1.24 .2138 -.00725 .03240ACDO:FEM| -.12345 .19669 -.63 .5302 -.50895 .26205ACDO:MAL| .01808 .13407 .13 .8927 -.24469 .28086ACDO:AGE| -.01336 .01241 -1.08 .2816 -.03767 .01096ACD0:AGE| -.00402 .00726 -.55 .5804 -.01825 .01022ACD1:AGE| -.00370 .00542 -.68 .4946 -.01433 .00693ACD2:AGE| .00094 .00451 .21 .8357 -.00791 .00978ACD3:AGE| -.00058 .00458 -.13 .8985 -.00956 .00839NPRS:FEM| 0.0 .....(Fixed Parameter).....NPRS:MAL| 0.0 .....(Fixed Parameter).....NPRS:AGE| 0.0 .....(Fixed Parameter).....NPR0:AGE| 0.0 .....(Fixed Parameter).....NPR1:AGE| 0.0 .....(Fixed Parameter).....NPR2:AGE| 0.0 .....(Fixed Parameter).....NPR3:AGE| 0.0 .....(Fixed Parameter).....CERT:FEM| 0.0 .....(Fixed Parameter).....CERT:MAL| 0.0 .....(Fixed Parameter).....CERT:AGE| 0.0 .....(Fixed Parameter).....CER0:AGE| 0.0 .....(Fixed Parameter).....CER1:AGE| 0.0 .....(Fixed Parameter).....CER2:AGE| 0.0 .....(Fixed Parameter).....CER3:AGE| 0.0 .....(Fixed Parameter).....CER0:FEM| 0.0 .....(Fixed Parameter).....CER0:MAL| 0.0 .....(Fixed Parameter).....CER4:AGE| 0.0 .....(Fixed Parameter).....CER5:AGE| 0.0 .....(Fixed Parameter).....CER6:AGE| 0.0 .....(Fixed Parameter).....CER7:AGE| 0.0 .....(Fixed Parameter).....CER8:AGE| 0.0 .....(Fixed Parameter).....MMOR:FEM| 0.0 .....(Fixed Parameter).....MMOR:MAL| 0.0 .....(Fixed Parameter).....MMOR:AGE| 0.0 .....(Fixed Parameter).....MMO0:AGE| 0.0 .....(Fixed Parameter).....MMO1:AGE| 0.0 .....(Fixed Parameter).....MMO2:AGE| 0.0 .....(Fixed Parameter).....MMO3:AGE| 0.0 .....(Fixed Parameter).....HMOR:FEM| 0.0 .....(Fixed Parameter).....HMOR:MAL| 0.0 .....(Fixed Parameter).....HMOR:AGE| 0.0 .....(Fixed Parameter).....HMO0:AGE| 0.0 .....(Fixed Parameter).....HMO1:AGE| 0.0 .....(Fixed Parameter).....HMO2:AGE| 0.0 .....(Fixed Parameter).....HMO3:AGE| 0.0 .....(Fixed Parameter)..... |Distns. of RPs. Std.Devs or limits of triangular.................... NsCAGED| 3.39632*** .19070 17.81 .0000 3.02255 3.77008NsACDOOR| 1.76933*** .09625 18.38 .0000 1.58069 1.95797 NsNPRS| 1.65796*** .07953 20.85 .0000 1.50209 1.81384 NsCERT1| .04752 .12448 .38 .7026 -.19645 .29150 NsCERT2| .51417*** .09248 5.56 .0000 .33292 .69542 NsMMORT| .13415 .12968 1.03 .3009 -.12002 .38831 NsHMORT| .85249*** .07967 10.70 .0000 .69633 1.00865--------+--------------------------------------------------------------------***, **, * ==> Significance at 1%, 5%, 10% level.Fixed parameter ... is constrained to equal the value orhad a nonpositive st.error because of an earlier problem.Model was estimated on Nov 02, 2022 at 07:54:32 PM-----------------------------------------------------------------------------
Parameter Matrix for Heterogeneity in Means.Matrix DELTA_RP is displayed in project window.Saved Individual Estimates of WTP in matrix WTP_I [ 955x5]Alternative Attribute Income/Cost Chosen CAGED PRICE Chosen ACDOORS PRICE Chosen NPRS PRICE Chosen CERT1 PRICE Chosen CERT2 PRICE(Saved absolute values. Check signs of coefficients.)+-------------------------------------------------------------------------+| Descriptive Statistics for Alternative ALTI1 || Utility Function | (CBS wt = 1.00000) | 2523.0 observs. || Coefficient | All 9550.0 obs.|that chose ALTI1 || Name Value Variable | Mean Std. Dev.|Mean Std. Dev. || ------------------- -------- | -------------------+------------------- || CAGED -3.4868 CAGED | .338 .473| .122 .328 || ACDOORS 2.3281 ACDOORS | .269 .443| .589 .492 || NPRS .0506 NPRS | .795 .404| .932 .251 || CERT1 .1925 CERT1 | .266 .442| .372 .483 || CERT2 .2220 CERT2 | .571 .495| .657 .475 || MMORT -.4868 MMORT | .234 .424| .187 .390 || HMORT -1.3133 HMORT | .442 .497| .319 .466 || PRICE -.1912 PRICE | 8.470 3.425| 8.673 3.088 |+-------------------------------------------------------------------------++-------------------------------------------------------------------------+| Descriptive Statistics for Alternative ALTI2 || Utility Function | (CBS wt = 1.00000) | 3076.0 observs. || Coefficient | All 9550.0 obs.|that chose ALTI2 || Name Value Variable | Mean Std. Dev.|Mean Std. Dev. || ------------------- -------- | -------------------+------------------- || CAGED -3.4868 CAGED | .507 .500| .239 .426 || ACDOORS 2.3281 ACDOORS | .363 .481| .618 .486 || NPRS .0506 NPRS | .895 .307| .955 .208 || CERT1 .1925 CERT1 | .275 .447| .422 .494 || CERT2 .2220 CERT2 | .490 .500| .467 .499 || MMORT -.4868 MMORT | .400 .490| .335 .472 || HMORT -1.3133 HMORT | .258 .438| .194 .396 || PRICE -.1912 PRICE | 8.561 3.497| 8.927 3.188 |+-------------------------------------------------------------------------++-------------------------------------------------------------------------+| Descriptive Statistics for Alternative NONE || Utility Function | (CBS wt = 1.00000) | 3951.0 observs. || Coefficient | All 9550.0 obs.|that chose NONE || Name Value Variable | Mean Std. Dev.|Mean Std. Dev. || ------------------- -------- | -------------------+------------------- || CAGED -3.4868 CAGED | .000 .000| .000 .000 || ACDOORS 2.3281 ACDOORS | .000 .000| .000 .000 || NPRS .0506 NPRS | .000 .000| .000 .000 || CERT1 .1925 CERT1 | .000 .000| .000 .000 || CERT2 .2220 CERT2 | .000 .000| .000 .000 || MMORT -.4868 MMORT | .027 .163| .016 .127 || HMORT -1.3133 HMORT | .000 .000| .000 .000 || PRICE -.1912 PRICE | .000 .000| .000 .000 || ASC -1.2560 ONE | 1.000 .000| 1.000 .000 |+-------------------------------------------------------------------------+
Elasticity wrt change of X in row choice on Prob[column choice]--------+--------------------------CAGED | ALTI1 ALTI2 NONE--------+-------------------------- ALTI1| .0784 -.0557 .0020 ALTI2| -.0522 .0635 .0042 NONE| .0000 .0000 .0000
More information about the Limdep
mailing list