# How can I implement Double-Exponential Smoothing for my data?

I have a time series problem that I'm dealing with and I'm looking for guidance on how to implement Double/Triple-Exponential Smoothing algorithm to predict my data. My problem is Taxi demand prediction. I have Jan-2015 data.

I have number of taxi pickups in $$10$$ minutes time-bins. $$i.e$$ each no_pickups entry is in 10 minutes bin and time-bins are from, $$\text{01-01-2015 00:00:00}$$ to $$\text{31-01-2015 00:00:00}.$$ There are about 4464 data-points because (31(days) * 24(hours) * 60(mins)) / 10(mins) = 4464 (time-bins)

How can I use these previous values to forecast it for Feb-2015?

I would really appreciate if you could tell me how to proceed with such problem and really looking forward to detailed answer so I can follow along. If anyone can help me out do let me know.

Example of data-set:

# Here's a sample data head.

time_bin                          n_pickups

(01-01-2015 00:00:00 - 01-01-2015 00:10:00]           50
(01-01-2015 00:10:00 - 01-01-2015 00:20:00]          108
(01-01-2015 00:20:00 - 01-01-2015 00:30:00]          160
.........                            ...
.........                            ...
.........                            ...
(31-01-2015 23:50:00 - 01-02-2015 00:00:00]           32