ETS function in package 'Forecast" in R, minimum data points

Question

For the ETS function in R, am I looking for the minimal number of data points for the forecast. I read both of the Hyndman papers (2002 & 2008) which are mentioned in the documentation, but I could not find a quantifiable value. In: http://robjhyndman.com/hyndsight/short-time-series/ he mentioned that it depends on the number of parameters . But at this point I am looking for a good source that can clarify the amount of data points needed in the ETS function. Can somebody help me with this?

Answer 1

Why don't you just try it and see what happens. You will find that the ets function will work with a 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, it is not fitting two parameters, but it is doing something reasonable given the information provided.

If you read the blog post you cite, I write "The only reasonable approach is to first check that there are enough observations to estimate the model, and then to test if the model performs well out-of-sample." You need more observations than parameters to actually estimate the model.