Does thetaf() in the forecast package in R detect seasonality? I had a question on the theta method in the forecast package in R. I tried using the AirPassenger example with thetaf(), but it just provided a trend line without seasonality in the forecast. Does thetaf() detect and forecast seasonality? 
This is the R code that I used:
library(forecast)
x.fit4 <- thetaf(AirPassengers, h=24)
plot(x.fit4)

 A: The theta method does not handle seasonality. If you read the help file for thetaf you will see that it says 

The theta method of Assimakopoulos and Nikolopoulos (2000) is
  equivalent to simple exponential smoothing with drift.

If you want a method that detects trend and seasonality, use either ets or auto.arima.
A: The theta model does not handle seasonality by itself, but the thetaf function in the forecast package does. If you read the help file for thetaf, you will see the following:

The theta method of Assimakopoulos and Nikolopoulos (2000) is equivalent to simple exponential smoothing with drift. This is demonstrated in Hyndman and Billah (2003).
The series is tested for seasonality using the test outlined in A&N. If deemed seasonal, the series is seasonally adjusted using a classical multiplicative decomposition before applying the theta method. The resulting forecasts are then reseasonalized.

(Source.)
If you read the theta model paper, they describe the seasonality test in more detail on page 524.
