Ossama Elshiewy, German Zenetti, and Yasemin Boztug, "Differences between Classical and Bayesian Estimates for Mixed Logit Models -- A Replication Study," Journal of Applied Econometrics, Vol. 32, No. 2, 2017, pp. 470-476. We provide all datasets and estimation codes to reproduce our results in ezb-files.zip. It contains all datasets as text files (ASCII files in DOS format), along with two .R program files. Unix/Linux users should use "unzip -a". The simulated datasets are also provided as RData files in ezb-rdata.zip. Since they are binary files, Unix/Linux users should not use the "-a" option. The information needed to make use of the empirical datasets is summarized below. The four simulated datasets are described as follows: The simulated panel data has N=300 individuals each with T=50 choice situations, and the simulated cross-sectional dataset has N=15,000 individuals with one choice situation each. In these datasets, we simulate the choice among three generic alternatives (X1, X2, X3) as a function of the variables Price and Promo. The true parameters for the simulated datasets are -1 for X3 and 0 for X2 as alternative-specific constants with X1 as the baseline category. The mean values of the random coefficients are -2 for Price and 1 for Promo. We simulate the correlation of the random coefficients as .75 in the high correlation scenario and .25 in the low correlation scenario. The datasets are named according to their classification Panel (_PANEL_) vs Cross-sectional (_CS_) and High Correlation (_cor_high_) vs. Low Correlation (_cor_low_). ### For convenient replication of our results in R we recommend the following instructions: Users who want to reproduce our results can download R for free from https://cran.r-project.org. The empirical datasets are then loaded from the R-Package "mlogit", which is also necessary to obtain the classical parameter estimation routine. For the Bayesian estimates, the R-Package "RSGHB" is necessary. As soon as users have successfully installed R and these packages, the two R-Codes can be run to load all datasets and reproduce our results. Users will find all necessary comments within these files: -- "R-Code_Classical_mlogit.R" has the estimation code to obtain all classical estimates. -- "R-Code_Bayes_RSGHB.R" has the estimation code to obtain all Bayesian estimates. After loading the packages and datasets into the Workload, a variable description for the empirical datasets (and their origin) can be obtained from R by using the command help(datasetname), for example help(Electricity) or help(Cracker). ### ** Descriptions of empirical datasets ** ********** ***Electricity Stated preference data for the choice of electricity suppliers panel data /number of observations/ : 4308 /observation/ : households /country/ : United States A dataframe containing : choice the choice of the individual, one of 1, 2, 3, 4, id the individual index, pfi fixed price at a stated cents per kWh, with the price varying over suppliers and experiments, for scenario i=(1, 2, 3, 4), cli the length of contract that the supplier offered, in years (such as 1 year or 5 years.) During this contract period, the supplier guaranteed the prices and the buyer would have to pay a penalty if he/she switched to another supplier. The supplier could offer no contract in which case either side could stop the agreement at any time. This is recorded as a contract length of 0 loci is the supplier a local company, wki is the supplier a well-known company todi a time-of-day rate under which the price is 11 cents per kWh from 8am to 8pm and 5 cents per kWh from 8pm to 8am. These TOD prices did not vary over suppliers or experiments: whenever the supplier was said to offer TOD, the prices were stated as above. seasi a seasonal rate under which the price is 10 cents per kWh in the summer, 8 cents per kWh in the winter, and 6 cents per kWh in the spring and fall. Like TOD rates, these prices did not vary. Note that the price is for the electricity only, not transmission and distribution, which is supplied by the local regulated utility. Source Hubert J, Train K (2001) "On the similarity of classical and Bayesian estimates of individual mean pathworths", Marketing Letters, 12 (3), 259-269. Kenneth Train's home page : http://elsa.berkeley.edu/~train/. ********** ***Cracker Choice of Brand for Crackers cross-sectional data /number of observations/ : 3292 /observation/ : individuals /country/ : United States A dataframe containing : id individuals identifiers choice one of sunshine, keebler, nabisco, private disp.z is there a display for brand z ? feat.z is there a newspaper feature advertisement for brand z ? price.z price of brand z Source Jain, Dipak C., Naufel J. Vilcassim and Pradeep K. Chintagunta (1994) "A random-coefficients logit brand-choice model applied to panel data", Journal of Business and Economics Statistics, 12(3), 317-328. Journal of Business Economics and Statistics web site : http://www.amstat.org/publications/jbes/. ********** ***Fishing Choice of Fishing Mode cross-sectional data /number of observations/ : 1182 /observation/ : individuals /country/ : United States A dataframe containing : mode recreation mode choice, one of : beach, pier, boat and charter price.beach price for beach mode price.pier price for pier mode price.boat price for private boat mode price.charter price for charter boat mode catch.beach catch rate for beach mode catch.pier catch rate for pier mode catch.boat catch rate for private boat mode catch.charter catch rate for charter boat mode income monthly income Source Herriges, J. A. and C. L. Kling (1999) "Nonlinear Income Effects in Random Utility Models", Review of Economics and Statistics, 81(1), 62-72. ###### Please address any questions to: Ossama Elshiewy oelshie [AT] gwdg.de Faculty of Economic Sciences University of Goettingen