bike {holiglm} | R Documentation |
Bike Sharing Dataset
Description
This data set contains the daily count of rented bikes from the the Capital Bikeshare system in Washington D.C., USA, for the years 2011 and 2012. The dataset is already prepared (correct types + factor encodings) for model building.
Format
A data.frame of dimension 731 x 12 containing daily data related to related bikes.
- dteday
a date vector giving the date of the rental.
- season
a factor with levels 'spring', 'summer', 'fall' and 'winter'.
- year
a factor with levels '2011' and '2012'.
- mnth
a factor with levels 'Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov' and 'Dec'.
- holiday
a boolean vector indicating if the day is a holiday.
- weathersit
a factor with levels 'good', 'neutral', 'bad' and 'very bad' giving the weather situation.
- temp
a numeric vector containing max-normalized temperature in Celsius with 41 as maximum.
- atemp
a numeric vector containing max-normalized feeling temperature in Celsius with 50 as maximum.
- hum
a numeric vector containing max-normalized humidity with 100 as maximum.
- windspeed
a numeric vector containing max-normalized windspeed with 67 as maximum.
- cnt
an integer vector containing counts of rented bikes.
Source
References
Fanaee-T, Hadi. (2013). Bike Sharing Dataset. UCI Machine Learning Repository.
Examples
data("bike")
hglm(formula = cnt ~ ., data=bike, family="poisson")