Basic Overview

Description

Forecasting function based on the autoregressive integrated moving average (ARIMA) model. More

Signature

ARIMA('Node', p, d, q, P, D, Q, m)

Parameters

  • Node: Node that contains the base data we want to base the forecasting on. It must contain at least one time level. Predictions is based on the time values in the node and it extends to the end of the horizon period. If the input data has a missing value in the time dimension the calculation fails. A time value is considered missing if the data has actual value(s) before and after the missing value. For example [2019, 2021] has missing value 2020, but [2020, 2021] or [2019, 2020] have no missing values.

  • p: the order (number of time lags) of the autoregressive model

  • d: the degree of differencing

  • q: the order of the moving-average model

  • P: autoregressive term for the seasonal part

  • D: differencing term for the seasonal part

  • Q: moving average term for the seasonal part

  • m: the number of periods in each season

Available from

3.6.0

  • The arguments p, d, q, P, D, Q, m can be nodes instead of single numbers, allowing to use different ARIMA arguments for individual time series in 'node'. Available since 3.7.0

Example

Input: Data from 2000-2011 (Passenger historic.xlsx)

Year

Month

MEASURE

2000

2000-01

112

...

...

...

2011

2011-12

432

Output: ARIMA('Passenger historic', 1, 1, 1, 1, 0, 1, 12)

Data from 2000-2020

Year

Month

MEASURE

2000

2000-01

112

...

...

...

2011

2011-12

432

2012

2012-01

450

...

...

...

2020

2020-12

861

Chart:



FAQs

Frequently asked questions can be found here

Contact 

You may contact the Valsight Customer Support via:

+49 30 46799042

support@valsight.com

Availability: Mon-Fri *, 9 AM to 5 PM (Berlin, Germany) .

*Except Public holidays in Berlin, Germany.