Exponential smoothing produces different values to R's forecast library

[MMenchero]

What happened + What you expected to happen

Using R's forecast package, the time series y = c(5,6,3,2,6,3,9) produces the following values when using Holt-Winter's method:

• alpha: 6.610696e-05
• fit: c(5.0000, 5.0001, 4.9999, 4.9997, 4.9998, 4.9997, 4.9999)

Using StatsForecast, which is supposed to be a mirror of the R package, we obtain an alpha of 1.00000000e-04 and different fitted values.

Versions / Dependencies

forecast R package version 8.16 StatsForecast version

Reproduction script

For R, use Holt-Winters function hw and for StatsForecast, use functions from Exponential Smoothing

Issue Severity

Low: It annoys or frustrates me.

[AFilipaVEL]

Hi!

I got the same issue (but with slightly higher severity :( )... are there any updates on this?

Thanks in advance!

[jmoralez]

Hey @AFilipaVEL, can you please provide a reproducible example?