Stated choices by 361 US households among four hypothetical electricity suppliers. Each household completed 8 to 12 choice tasks.
This data can be used to model the choice between electricity suppliers based on price, contract length, whether the supplier is local or well-known, and whether time-of-day or seasonal rates are offered.
Format
A tibble with 4308 rows and 27 columns:
- household [
integer] The household identifier.
- occasion [
integer] The choice occasion within a household.
- choice [
character] The chosen supplier, from
"1"to"4".- price_1–price_4 [
numeric] The fixed electricity price in US-dollar cents per kWh for each supplier.
- contract_1–contract_4 [
numeric] The contract length in years for each supplier.
- local_1–local_4 [
logical] Whether each supplier is local.
- known_1–known_4 [
logical] Whether each supplier is well-known.
- timeOfDay_1–timeOfDay_4 [
logical] Whether each supplier offers a time-of-day rate.
- seasonal_1–seasonal_4 [
logical] Whether each supplier offers a seasonal rate.
Source
Adapted from Electricity in the
mlogit package.
The study is described by Huber and Train (2001).
References
Huber J, Train K (2001). “On the similarity of classical and Bayesian estimates of individual mean partworths.” Marketing Letters, 12(3), 259–269. doi:10.1023/A:1011120928698 .