Machine Learning
About 1 min
Machine Learning
AR
Usage
This function is used to learn the coefficients of the autoregressive models for a time series.
Name: AR
Input Series: Only support a single input numeric series. The type is INT32 / INT64 / FLOAT / DOUBLE.
Parameters:
p
: The order of the autoregressive model. Its default value is 1.
Output Series: Output a single series. The type is DOUBLE. The first line corresponds to the first order coefficient, and so on.
Note:
- Parameter
p
should be a positive integer. - Most points in the series should be sampled at a constant time interval.
- Linear interpolation is applied for the missing points in the series.
Examples
Assigning Model Order
Input Series:
+-----------------------------+---------------+
| Time|root.test.d0.s0|
+-----------------------------+---------------+
|2020-01-01T00:00:01.000+08:00| -4.0|
|2020-01-01T00:00:02.000+08:00| -3.0|
|2020-01-01T00:00:03.000+08:00| -2.0|
|2020-01-01T00:00:04.000+08:00| -1.0|
|2020-01-01T00:00:05.000+08:00| 0.0|
|2020-01-01T00:00:06.000+08:00| 1.0|
|2020-01-01T00:00:07.000+08:00| 2.0|
|2020-01-01T00:00:08.000+08:00| 3.0|
|2020-01-01T00:00:09.000+08:00| 4.0|
+-----------------------------+---------------+
SQL for query:
select ar(s0,"p"="2") from root.test.d0
Output Series:
+-----------------------------+---------------------------+
| Time|ar(root.test.d0.s0,"p"="2")|
+-----------------------------+---------------------------+
|1970-01-01T08:00:00.001+08:00| 0.9429|
|1970-01-01T08:00:00.002+08:00| -0.2571|
+-----------------------------+---------------------------+