kalmanMinMax (kalman based minima maxima detection)

The kalmanMinMax function uses the kalman filter to smooth the time series values in the given window size, and then determine the maxima and minima of that set of values.

Input parameters

Parameter	Required/Optional	Description
Variable	Required	The time series value to be considered for minima maxima detection.
Q	Required	The standard deviation of the process noise.
R	Required	The standard deviation of the measurement noise.
Window Size	Required	The number of values to be considered for smoothing and determining the extremes.
Extrema Type	Required	This can be `min`, `max` or `minmax`.

Output parameters

Parameter	Name	Description
min or max value	The variable name specified in the input parameter.	The value of the `min` or `max` point.
extrema Type	extrema Type	Indicates whether the returned value is a `min` value or a `max` value.

Examples

Minimum values

The following returns the maximum values for a set of price values.

from inputStream#timeseries:kalmanMinMax(price, 0.000001,0.0001, 25, 'min')
select *
insert into outputStream;

Maximum values

The following returns the minimum values for a set of price values.

from inputStream#timeseries:kalmanMinMax(price, 0.000001,0.0001, 25, max)
select *
insert into outputStream;

Minimum and maximum values

The following returns both the minimum values and the maximum values for a set of price values.

from inputStream#timeseries:kalmanMinMax(price, 0.000001,0.0001, 25, 'minmax')
select *
insert into outputStream;