Forecast

Siddhi enables users to forecast future events using linear regression on real time data streams. The forecast function takes in a dependent event stream (Y), an independent event stream (X) and a user specified next X value, and returns the forecast Y value based on the regression equation of the historical data.

Input Parameters

Parameter		Description
Calculation Interval	Optional	The frequency of the regression calculation. Default value: 1 (i.e. at every event)
Batch Size	Optional	The maximum number of events that should be used for a regression calculation. Default value: 1,000,000,000 events
Confidence Interval	Optional	Confidence Interval to be used for regression calculation Default value: 0.95
Next X Value	Required	Value to be used to forecast the Y value. This can be a constant, or an expression (eg:- x+5)
Y Stream	Required	Data stream of the dependent variable
X Stream	Required	Data stream of the independent variable

Output Parameters

Parameter	Name	Description
Forecast Y	forecastY	Forecast Y value based on next X and regression equation.
Standard Error	stdError	Standard Error of the Regression Equation
β coefficients	beta0, beta1	β coefficients of the simple linear regression
Input Stream Data	Name given in the input stream	All items sent in the input stream

Example

The following query submits an expression to be used as the next X value (X+2) a dependent input stream (Y) and an independent input stream (X) that will be used to perform linear regression between Y and X streams, and compute the forecast Y value based on the next X value given by the user.

from StockExchangeStream#transform.timeseries:forecast(X+5, Y, X)

select *

insert into StockForecaster

When executed, the above query will return the forecast Y value based on the regression equation established using the Y stream and the X stream and the standard error of the regression equation (ε), β coefficients and all the items available in the input stream.