Auto ARIMA

Auto ARIMA

Auto ARIMA is located under Forecasting > Modeling > Auto ARIMA. Use the drag-and-drop method (or double-click on the node) to use the algorithm in the canvas. Click the algorithm to view and select different properties for analysis.


Properties of Auto ARIMA

The properties of Auto ARIMA are shown in the figure below.



The table give below shows the different properties of Auto ARIMA.

Field

Description

Remark


It helps execute the node.

-


It helps to explore the successful node.

-


It displays the following options in the list view.

  • Run till node
  • Run from node
  • Publish as a model
  • Publish code

-

Task Name

It is the name of the task selected on the workbook canvas.

You can click the text field to edit or modify the task name as required.

Time ID Variable

It allows you to select the interval type variable for which we need to process the dependent or target variable's values.

  • Only one data field can be selected.
  • If selected, only an interval-type variable should be selected.
  • Variables with numerical values are not available.

Target Variable

It allows you to select the experimental or predictor variable(s).

  • Only one data field can be selected.
  • Variables with only numerical values are available.

Group By

It allows you to select the variable for which you want to group the data.

  • Multiple data fields can be selected.
  • Only categorical type variables are available.

ADVANCED



Number of Periods for Forecasting

It allows you to select the time /date variable

Only one variable can be selected

Interval

It allows to select the interval for the accumulation of data in the accumulation test.

Available options are –

  • Day
  • Week
  • Month
  • Quarter
  • Year
    The default value is Month.

Start p

It is the starting value for the order of Autoregressive model.

The value must be positive.
The default value is 2.

Start q

It is the starting value for the order of Moving Average model.

The value must be positive.
The default value is 5.

d

It is the order of first differencing

The value must be positive or None.

Max p

The maximum value for the order of the Autoregressive model

The value must be positive.
The value should be greater than or equal to start p.
The default value is 5.

Max q

The maximum value for the order of the Moving averages model

The value must be positive.
The value should be greater than or equal to start q.
The default value is 5.

Max d

It is the maximum number of non-seasonal differences

The value must be positive.
The value should be greater than or equal to d.
The default value is 2.

Start P

It is the starting value of P, the order of the AR portion of the seasonal model.

The default value is 1.

Start Q

It is the starting value of Q, the order of the MA portion of the seasonal model.

The default value is 1.

D

It is the order of seasonal differencing

The value must be positive or None.

Max P

It is the maximum value of P

The value must be a positive integer greater than or equal to Start P.
The default value is 2.

Max Q

It is the maximum value of Q

The value must be a positive integer greater than or equal to Start Q.
The default value is 2.

Max D

It is the maximum value of D

The value must be a positive integer greater than or equal to Start P.
The default value is 2.

Max order

It is the maximum p+q+P+Q if the model selection is not stepwise

The default value is 5

m

It is the period for seasonal differencing

The default value is 1.

Seasonal

It allows you to fit the seasonal ARIMA model.

The following options are available:

  • True
  • False
    The default value is True

Stationary

It allows you to select the autoregressive parameters to correspond to a stationarity process.

Available options are

  • True
  • False
    The default value is False

Information Criterion

It Selects the best ARIMA model.

The following options are available

  • aic
  • bic
  • hqix
  • oob

Alpha

It is the level of test for testing significance

The default value is 0.05.

Test

It is the type of unit root test used to detect stationarity.

The default value is KPSS

Seasonal Test

It determines which seasonal unit root test is used.

The default value is OCSB.

Stepwise

It allows the usage of stepwise algorithm to identify the optimal model parameters

The following options are available

  • True
  • False
    The default value is True.

Method

It determines which solver is used

The following options are available

  • newton
  • nm
  • bfgs
  • lbfgs
  • powell
  • cg
  • ncg
  • basinhopping
    The default value is lbfgs.

Trend

It is the trend parameter

-

Suppress Warnings

It allows to suppress the error coming from ARIMA.

-

Error Action

It allows error handling in case the ARIMA model cannot be fitted

The following options are available

  • warm
  • raise
  • ignore
  • trace
    The default value is warm

Trace

It allows us to print the status of the fits.

-

Random State

Ensures replicable testing and results.

-

Scoring

It is the metric to use for scoring the out-of-sample data while performing validation

The following options are available

  • mse
  • mae
    By default, the value is mse.

With Intercept

It allows you to include an intercept form.

-

Example of Auto ARIMA

Consider the Air Passengers dataset. The snippet of input data is shown below.

We apply the Train Test Split to the input data and then use the Auto ARIMA to the Train Test Split node.


We select the TravelDate as the Time ID Variable and Passengers as Target variable.


Further, the result page displays,

  • Forecasting chart with predicted value.
  • Trained Model Parameters
  • Accuracy Parameters



    • Related Articles

    • ARIMA

      ARIMA is located under Forecasting > Modeling > ARIMA. Use the drag-and-drop method (or double-click on the node) to use the algorithm in the canvas. Click the algorithm to view and select different properties for analysis. Properties of ARIMA The ...

    • Train Test Split

      Train Test Split is located under Forecasting ( ) in Data Preparation, in the left task pane. Use the drag-and-drop method to use the algorithm in the canvas. Click the algorithm to view and select different properties for analysis. Refer to ...

    • Components in Forecasting

      Forecasting deals with the analysis and detection of trends in the time-series data. The components of forecasting are, Data Exploration: Data Exploration is used to explore the time-series data. It helps in identifying the underlying parameters ...

    • Features of AutoML Wizard

      The Auto ML Wizard recommends the best fit model for a selected dataset, This recommendation is especially helpful if you do not know which algorithm to use for the selected dataset. Features: Auto-creation of a workbook Import, search or choose a ...

    • AutoML

      Introduction to AutoML Wizard AutoML functionality in Rubiscape for Predictive Analytics, you can create, test, deploy, and maintain your predictive models easily and instantly. The Auto ML functionality offers interactive data exploration and makes ...