Arima

Syntax

Arima(Input: ITimeSeries,
      Period: IMsPeriod,
      NotSeasonalIAR: Integer,
      NotSeasonalIMA: Integer,
      NotSeasonalDIFF: Integer,
      ConstantValue: Variant,
      SeasonalAR: Integer,
      SeasonalMA: Integer,
      SeasonalDiff: Integer,
      SeasonalPeriod: Integer,
      MaxIteration: Integer,
      Precision: Double,
      Casewise: MsCasewise)

Parameters

Input. Output variable.

Period. Period, at which the method is calculated. If the parameter value is Null, the method is calculated at the entire time period.

NotSeasonalIAR. Non-seasonal autoregression order.

NotSeasonalIMA. Non-seasonal moving average order.

NotSeasonalDIFF. Non-seasonal difference order.

ConstantValue. Constant used in calculations.

SeasonalAR. Seasonal regression order. Optional parameter. The parameter is set to 0 by default.

SeasonalMA. Seasonal moving average order. Optional parameter. The parameter is set to 0 by default.

SeasonalDiff. Seasonal difference order. Optional parameter. The parameter is set to 1 by default.

SeasonalPeriod. Duration of seasonal period. Optional parameter. The parameter is set to 0 by default.

MaxIteration. Maximum number of iterations, in which the optimal decision must be found. Optional parameter. The parameter is set to 500 by default.

Precision. Calculation accuracy. Optional parameter. The parameter is set to 0.0001 by default.

Casewise. Missing data treatment method. Optional parameter. The parameter is set to MsCasewise.No by default - missing data treatment is not used.

Description

It models series values using the ARIMA method.

Comments

ConstantValue. The constant value can be determined by the user or estimated automatically. Use the Estimate function, to estimate values automatically. If the model must be calculated without constant, use the None function.

NotSeasonalIAR, NotSeasonalIMA, SeasonalAR, SeasonalMA. Parameters are set in the string view. Specify the numbers or ranges of autoregression or moving average orders separated by commas. The range of autoregression or moving average orders is specified via the sign -. For example: SeasonalAR = "1-3,5".

NotSeasonalIMA, SeasonalMA. If the order of the moving average (seasonal/non-seasonal) is set, the back-casting can be used on estimate of its coefficients. Back-casting parameters are set once in the NotSeasonalIMA parameter and used for all types of moving average. By default the back-casting is used. If the back-casting should be disabled, the NotSeasonalIMA parameter must contain the "backcast.No" string. For example: NotSeasonalIMA = "1-4;backcast.No".

Example

Formula	Result	Application
= Arima({Brazil\|BCA[t]}, SetPeriod("2001", "2016"), "1-3", "1,4;backcast.No", 0, Estimate)	For the Brazil\|BCA series the ARIMA method is calculated by the following parameters: order of non-seasonal autoregression is in the range 1-3, order of non-seasonal moving average is 1.4, back-casting is not used to estimate seasonal average coefficients, non-seasonal difference order is 0, constant value is estimated automatically, using the Estimate function. Calculation is made without missing data treatment for the period from 2001 to 2016.	It can be used in formulas of calculated series of time series database and in formulas of attribute-based models of modeling container.
= Arima(X1,Null,"","",1,2.7,"","", 1,0,500,0.0001, MsCasewise.Yes)	For the X1 factor ARIMA method is calculated by the following parameters: orders of non-seasonal autoregression and non-seasonal moving average are not assigned, order of non-seasonal difference is 1, constant value is 2.7, calculation is executed for the whole period without considering missing values, missing data will be treated with the Casewise method.	It can be used in model variable-based formulas of modeling container.