# Simple exponential smoothing, how extract α and ^ ℓ 0?

HI all,
I have a few questions:

1. Is there a code to extract the optimal smoothing constant once you have run the ETS model?
(the example shows α =0.84 )

2. Is there a code to extract the ℓ0 once you have run the ETS model?
( ℓ0 for in the year 1959 , (in the example is = 39.54))

3. How do you get to 39.54? ( I don´t understand how the calculation is done and why)

``````# Estimate parameters
fit <- algeria_economy %>%
*model(ETS(Exports ~ error("A") + trend("N") + season("N")))*
fc <- fit %>%
forecast(h = 5)
``````

Thanks in advance!

The `fpp3` library doesn't include the `algeria_economy` dataset, but `aus_airpassengers` is similar

``````library(fpp3)
fit <- aus_airpassengers %>%
model(ETS(Passengers ~ error("A") + trend("N") + season("N")))
fc <- fit %>%
forecast(h = 5)
report(fit)
#> Series: Passengers
#> Model: ETS(A,N,N)
#>   Smoothing parameters:
#>     alpha = 0.9998998
#>
#>   Initial states:
#>      l[0]
#>  7.319845
#>
#>   sigma^2:  6.3296
#>
#>      AIC     AICc      BIC
#> 271.6389 272.1970 277.1894
``````

Chapter 8 gives more detail of calculations in the context of state models.

1 Like

many thanks! report is exactly what I needed

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.