Sarimax

Name : Sarimax

Description: SARIMAX (Seasonal AutoRegressive Integrated Moving Average with eXogenous regressors) is a time series forecasting model that extends ARIMA by incorporating seasonal effects and external variables. It is used to capture complex patterns in data with both trend and seasonality

Input : Dataframe - The latest version of uploaded data

Output : Dataframe - Link to the saved Dataframe which is modified as per the function

Model - Model that is fit on the training data and Summary of the actions performed

Mandatory and Non-mandatory/Advanced hyperparameters required for the function are listed below

Mandatory
Non-mandatory/Advanced hyperparameters

target_col

index_column

modelname

columns

p

q

d

p

Q

D

s

trend

Save As New Table

Details of Hyperparameters are provided below

Name
Description
Accepted range/ values
Display Name

target_col

Target column

Columns in the dataframe

Target column

modelname

It is the name of the model required for saving

String

Name of the model

index_column

Column that can be used as index column for forecasting

Columns in the dataframe

Index column

columns

Numerical columns to be used for building models

Columns in the dataframe

Columns to select

p

Number of lag observations or autoregressive terms in the model

Integer in the range [1,inf)

p

q

Number of times that the raw observations are differenced

Integer in the range [1,inf)

q

d

Size of the moving average window

Integer in the range [1,inf)

d

P

Number of lagged seasonal observations or autoregressive terms in the model.It indicates the number of previous time steps to consider for predicting the current time step

Integer in the range [1,inf)

p

Q

Number of times that the raw seasonal observations are differenced

Integer in the range [1,inf)

Q

D

Size of the moving average window

Integer in the range [1,inf)

D

s

Parameter that indicates number of periods in a season

Integer in the range [2,inf)

S

trend

Parameter controlling the deterministic trend polynomial

c,n,t,nt,None

Trend

Save As New Table

Option to choose whether the dataframe is to be saved as a new table

-

Last updated