Stochastic PARMA Model Type

Units:	-
Default Value:	-1
Validation Rule:	In (-1,0,1)
Description:	PARMA Model Type (-1 = None, 0 = SARIMA model, 1 = PARMA model)
Detail:

This option controls the use of PARMA during the constructions of samples for the scenario tree. The option can take the following values:

None (value = -1): PARMA sampling will be turned off with this setting.
SARIMA (value = 0): A seasonal autoregressive integrated moving average (SARIMA) model is fitted to historical time series data and used for generating the samples for the scenario tree.
PARMA (value = 1): A periodic autoregressive (PAR) model is fitted to historical time series data and used for generating the samples for the scenario tree.

Seasonality in a time series is a regular pattern of changes that repeats over multiple time periods. For example, in monthly river flow data there is seasonality where in summer months the flow tends to be high and in winter months the flow is relatively low. While both SARIMA and PARMA models are capable of modelling a wide range of data with seasonal or periodic characteristics, evidence shows that PARMA is more suitable than SARIMA in modeling series such as natural inflow. Users are recommended to conduct time series analysis for the model selection before applying PARMA sampling to the specific property.