ApolloChoiceModelling forum

Hi
I use r 4.3.2. on RStudio
I installed apollo 0.3.1, and 0.2.9.
After running library(apollo), it is loaded but with the next message:

> library(apollo)


. ,,
, ,,
,,,,,, , ,,
, ,, , ,,,,.
,, , ,, ,,,,,, ,,, // //
, ,,,. ,,,,,. ,, //// // //
,, ,,,,,. ,, // // ////// ///// // // /////
,,, ,, , // // / // // // // // // //
,, , //////// / // // // // // // //
,, ,, // // / /// // // // // // //
, // // ///// /// // // ///
//
//

Apollo 0.3.1
http://www.ApolloChoiceModelling.com
See url for a detailed manual, examples and a user forum.
Sign up to the user forum to receive updates on new releases.

Please cite Apollo in all written material you produce:
Hess S, Palma D (2019). "Apollo: a flexible, powerful and customisable
freeware package for choice model estimation and application." Journal
of Choice Modelling, 32. doi:10.1016/j.jocm.2019.100170
Warning message:
In .recacheSubclasses(def@className, def, env) :
undefined subclass "ndiMatrix" of class "replValueSp"; definition not updated

Also (I do not know if it is related) I tried to run the example "MDCEV model on time use data, alpha-gamma profile with outside good and socio-demographics" (I ran the example exactly as appears in the apollo examples page) and while estimating, I get:

Calculating log-likelihood at equal shares (LL(0)) for applicable models...
Calculating log-likelihood at observed shares from estimation data (LL(c)) for applicable models...
Calculating LL of each model component...
Computing covariance matrix using numerical methods (numDeriv).
0%....25%....50%....75%....100%
WARNING: Singular Hessian, cannot calculate s.e.
Hessian written to output/MDCEV_with_outside_good_hessian.csv
WARNING: Some eigenvalues of the Hessian are positive, indicating convergence to a saddle point!
Computing score matrix...
Your model was estimated using the BGW algorithm. Please acknowledge this by citing Bunch et al. (1993) - DOI 10.1145/151271.151279
Warning messages:
1: In log((inputs$continuousChoice[[j]]/inputs$gamma[[j]]) + 1) :
NaNs produced
2: In log(inputs$continuousChoice[[j]] + inputs$gamma[[j]]) :
NaNs produced
3: In log((inputs$continuousChoice[[j]]/inputs$gamma[[j]]) + 1) :
NaNs produced
4: In log(inputs$continuousChoice[[j]] + inputs$gamma[[j]]) :
NaNs produced
5: In log((inputs$continuousChoice[[j]]/inputs$gamma[[j]]) + 1) :
NaNs produced
6: In log(inputs$continuousChoice[[j]] + inputs$gamma[[j]]) :
NaNs produced
7: In log((inputs$continuousChoice[[j]]/inputs$gamma[[j]]) + 1) :
NaNs produced
8: In log(inputs$continuousChoice[[j]] + inputs$gamma[[j]]) :
NaNs produced
> apollo_modelOutput(model)
Model run by haort using Apollo 0.3.1 on R 4.3.2 for Windows.
Please acknowledge the use of Apollo by citing Hess & Palma (2019)
DOI 10.1016/j.jocm.2019.100170
www.ApolloChoiceModelling.com

Model name : MDCEV_with_outside_good
Model description : MDCEV model on time use data, alpha-gamma profile with outside good and socio-demographics
Model run at : 2023-11-22 17:50:59.58616
Estimation method : bgw
Model diagnosis : Relative function convergence
Optimisation diagnosis : Saddle point found
hessian properties : Some eigenvalues are positive and others negative
maximum eigenvalue : 0
reciprocal of condition number : not calculated (Hessian is not negative definite)
Number of individuals : 447
Number of rows in database : 2826
Number of modelled outcomes : 2826

Number of cores used : 1
Model without mixing

LL(start) : -21594.92
LL at equal shares, LL(0) : NA
LL at observed shares, LL(C) : NA
LL(final) : -15007.36
Rho-squared vs equal shares : Not applicable
Adj.Rho-squared vs equal shares : Not applicable
Rho-squared vs observed shares : Not applicable
Adj.Rho-squared vs observed shares : Not applicable
AIC : 30044.72
BIC : 30133.92

Estimated parameters : 15
Time taken (hh:mm:ss) : 00:00:33.44
pre-estimation : 00:00:1.19
estimation : 00:00:14.75
post-estimation : 00:00:17.5
Iterations : 36

Unconstrained optimisation.

Estimates:
Estimate s.e. t.rat.(0) Rob.s.e. Rob.t.rat.(0)
alpha_base -911.5129 NA NA NA NA
gamma_work 4.8990 NA NA NA NA
gamma_school 3.0985 NA NA NA NA
gamma_shopping 0.4264 NA NA NA NA
gamma_private 0.6200 NA NA NA NA
gamma_leisure 2.0958 NA NA NA NA
delta_work -3.7172 NA NA NA NA
delta_school -7.4183 NA NA NA NA
delta_shopping -3.8044 NA NA NA NA
delta_private -4.2786 NA NA NA NA
delta_leisure -3.4001 NA NA NA NA
delta_work_FT 1.3248 NA NA NA NA
delta_work_wknd -2.8609 NA NA NA NA
delta_school_young 2.3442 NA NA NA NA
delta_leisure_wknd 0.2949 NA NA NA NA
sig 1.0000 NA NA NA NA

Hi

thanks for alerting us to this. With the new optimiser (BGW), the model correctly identifies an issue with the estimation of alpha_base, as this tends to -inf. If you now fix this to a large negative value by settingt the value in apollo_beta to e.g. the solution it goes to, and including the parameter name in apollo_fixed, everything should work

Thanks

Stephane & David