forecasting - ETS function in package 'Forecast" in R, minimum data points -
for ets function in r, looking minimal number of data points forecast. read both of hyndman papers (2002 & 2008) mentioned in documentation, not find quantifiable value. in: http://robjhyndman.com/hyndsight/short-time-series/ mentioned depends on number of parameters . @ point looking source can clarify amount of data points needed in ets function. can me this?
why don't try , see happens. find ets
function work single observation:
> 1 %>% ets %>% forecast point forecast lo 80 hi 80 lo 95 hi 95 2 1 1 1 1 1 3 1 1 1 1 1 4 1 1 1 1 1 5 1 1 1 1 1 6 1 1 1 1 1 7 1 1 1 1 1 8 1 1 1 1 1 9 1 1 1 1 1 10 1 1 1 1 1 11 1 1 1 1 1
of course, not fitting 2 parameters, doing reasonable given information provided.
if read blog post cite, write "the reasonable approach first check there enough observations estimate model, , test if model performs out-of-sample." need more observations parameters estimate model.
Comments
Post a Comment