dstsCV: Double seasonal exponential time series cross validation
In placeboo/amplify: Forecasting Daily Time Seriers with Double Seasonal Patterns

The cross-validation procedure based on a rolling forecasting.

dstsCV(
  train.y,
  valid.y,
  s1 = 7,
  s2 = 7 * 52,
  kfold = 4,
  par = list(alpha, beta, gamma, omega, phi)
)

`train.y`	A numeric vector for training.
`valid.y`	A numeric vector for testing.
`s1`	Period of the shorter seasonal period.
`s2`	Period of the longer seasonal period.
`kfold`	The number of subsets by spliting `valid.y`
`par`	The list of parameters for Double seasonal exponential time series model