smooth (version 1.9.0)

iss: Intermittent State Space

Description

Functin calculates the probability for intermittent state space model. This is needed in order to forecast intermittent demand using other functions.

Usage

iss(data, intermittent = c("none", "fixed", "croston", "tsb", "sba"),
  h = 10, holdout = FALSE, model = NULL, persistence = NULL)

Arguments

data

Either numeric vector or time series vector.

intermittent

Type of method used in probability estimation. Can be "none" - none, "fixed" - constant probability, "croston" - estimated using Croston, 1972 method and "TSB" - Teunter et al., 2011 method., "sba" - Syntetos-Boylan Approximation for Croston's method (bias correction) discussed in Syntetos and Boylan, 2005.

h

Forecast horizon.

holdout

If TRUE, holdout sample of size h is taken from the end of the data.

model

Type of ETS model used for the estimation. Normally this should be either "ANN" or "MNN".

persistence

Persistence vector. If NULL, then it is estimated.

Value

The object of class "iss" is returned. It contains following list of values:

• fitted - fitted values of the constructed model;
• states - values of states (currently level only);
• forecast - forecast for h observations ahead;
• variance - conditional variance of the forecast;
• logLik - likelihood value for the model
• nParam - number of parameters used in the model;
• residuals - residuals of the model;
• C - vector of all the parameters.
• actuals - actual values of probabilities (zeroes and ones).

Details

The function estimates probability of demand occurance, using one of the ETS state-space models.

References

• Teunter R., Syntetos A., Babai Z. (2011). Intermittent demand: Linking forecasting to inventory obsolescence. European Journal of Operational Research, 214, 606-615.
• Croston, J. (1972) Forecasting and stock control for intermittent demands. Operational Research Quarterly, 23(3), 289-303.
• Syntetos, A., Boylan J. (2005) The accuracy of intermittent demand estimates. International Journal of Forecasting, 21(2), 303-314.

See Also

ets, forecast, es

Examples

    y <- rpois(100,0.1)
    iss(y, intermittent="t")

    iss(y, intermittent="c", persistence=0.1)