Java 类名：com.alibaba.alink.operator.batch.timeseries.ArimaBatchOp
Python 类名：ArimaBatchOp

功能介绍

给定分组，对每一组的数据进行Arima时间序列预测，给出下一时间段的结果。

算法原理

Arima全称为自回归积分滑动平均模型(Autoregressive Integrated Moving Average Model,简记ARIMA)，是由博克思(Box)和詹金斯(Jenkins)于70年代初提出一著名时间序列预测方法，所以又称为box-jenkins模型、博克思-詹金斯法.
Arima 详细介绍请见链接 https://en.wikipedia.org/wiki/Autoregressive_integrated_moving_average

使用方式

参考文档 https://www.yuque.com/pinshu/alink_guide/xbp5ky

参数说明

| 名称 | 中文名称 | 描述 | 类型 | 是否必须？ | 取值范围 | 默认值 | | —- | —- | —- | —- | —- | —- | —- |

| order | 模型(p, d, q) | 模型(p, d, q) | int[] | ✓ | | |

| predictionCol | 预测结果列名 | 预测结果列名 | String | ✓ | | |

| valueCol | value列，类型为MTable | value列，类型为MTable | String | ✓ | 所选列类型为 [M_TABLE] | |

| estMethod | 估计方法 | 估计方法 | String | | “Mom”, “Hr”, “Css”, “CssMle” | “CssMle” |

| predictNum | 预测条数 | 预测条数 | Integer | | | 1 |

| predictionDetailCol | 预测详细信息列名 | 预测详细信息列名 | String | | | |

| reservedCols | 算法保留列名 | 算法保留列 | String[] | | | null |

| seasonalOrder | 季节模型(p, d, q) | 季节模型(p, d, q) | int[] | | | null |

| seasonalPeriod | 季节周期 | 季节周期 | Integer | | [1, +inf) | 1 |

| numThreads | 组件多线程线程个数 | 组件多线程线程个数 | Integer | | | 1 |

代码示例

Python 代码

  1. from pyalink.alink import *
  3. import pandas as pd
  5. useLocalEnv(1)
  7. import time, datetime
  8. import numpy as np
  9. import pandas as pd
  11. data = pd.DataFrame([
  12.             [1,  datetime.datetime.fromtimestamp(1), 10.0],
  13.             [1,  datetime.datetime.fromtimestamp(2), 11.0],
  14.             [1,  datetime.datetime.fromtimestamp(3), 12.0],
  15.             [1,  datetime.datetime.fromtimestamp(4), 13.0],
  16.             [1,  datetime.datetime.fromtimestamp(5), 14.0],
  17.             [1,  datetime.datetime.fromtimestamp(6), 15.0],
  18.             [1,  datetime.datetime.fromtimestamp(7), 16.0],
  19.             [1,  datetime.datetime.fromtimestamp(8), 17.0],
  20.             [1,  datetime.datetime.fromtimestamp(9), 18.0],
  21.             [1,  datetime.datetime.fromtimestamp(10), 19.0]
  22. ])
  24. source = dataframeToOperator(data, schemaStr='id int, ts timestamp, val double', op_type='batch')
  26. source.link(
  27.         GroupByBatchOp()
  28.             .setGroupByPredicate("id")
  29.             .setSelectClause("id, mtable_agg(ts, val) as data")
  30.         ).link(ArimaBatchOp()
  31.             .setValueCol("data")
  32.             .setOrder([1, 2, 1])
  33.             .setPredictNum(12)
  34.             .setPredictionCol("predict")
  35.         ).print()

Java 代码

  1. package com.alibaba.alink.operator.batch.timeseries;
  3. import org.apache.flink.types.Row;
  5. import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
  6. import com.alibaba.alink.operator.batch.sql.GroupByBatchOp;
  7. import com.alibaba.alink.testutil.AlinkTestBase;
  8. import org.junit.Test;
  10. import java.sql.Timestamp;
  11. import java.util.Arrays;
  12. import java.util.List;
  14. public class ArimaBatchOpTest extends AlinkTestBase {
  16.     @Test
  17.     public void test() throws Exception {
  18.         List <Row> mTableData = Arrays.asList(
  19.             Row.of(1, new Timestamp(1), 10.0),
  20.             Row.of(1, new Timestamp(2), 11.0),
  21.             Row.of(1, new Timestamp(3), 12.0),
  22.             Row.of(1, new Timestamp(4), 13.0),
  23.             Row.of(1, new Timestamp(5), 14.0),
  24.             Row.of(1, new Timestamp(6), 15.0),
  25.             Row.of(1, new Timestamp(7), 16.0),
  26.             Row.of(1, new Timestamp(8), 17.0),
  27.             Row.of(1, new Timestamp(9), 18.0),
  28.             Row.of(1, new Timestamp(10), 19.0)
  29.         );
  31.         MemSourceBatchOp source = new MemSourceBatchOp(mTableData, new String[] {"id", "ts", "val"});
  33.         source.link(
  34.             new GroupByBatchOp()
  35.                 .setGroupByPredicate("id")
  36.                 .setSelectClause("mtable_agg(ts, val) as data")
  37.         ).link(new ArimaBatchOp()
  38.             .setValueCol("data")
  39.             .setOrder(new int[] {1, 2, 1})
  40.             .setPredictNum(12)
  41.             .setPredictionCol("predict")
  42.         ).print();
  43.     }
  44. }

运行结果

| id | data | predict | | —- | —- | —- |

| 1 | {“data”:{“ts”:[“1970-01-01 08:00:00.001”,”1970-01-01 08:00:00.002”,”1970-01-01 08:00:00.003”,”1970-01-01 08:00:00.004”,”1970-01-01 08:00:00.005”,”1970-01-01 08:00:00.006”,”1970-01-01 08:00:00.007”,”1970-01-01 08:00:00.008”,”1970-01-01 08:00:00.009”,”1970-01-01 08:00:00.01”],”val”:[10.0,11.0,12.0,13.0,14.0,15.0,16.0,17.0,18.0,19.0]},”schema”:”ts TIMESTAMP,val DOUBLE”} | {“data”:{“ts”:[“1970-01-01 08:00:00.011”,”1970-01-01 08:00:00.012”,”1970-01-01 08:00:00.013”,”1970-01-01 08:00:00.014”,”1970-01-01 08:00:00.015”,”1970-01-01 08:00:00.016”,”1970-01-01 08:00:00.017”,”1970-01-01 08:00:00.018”,”1970-01-01 08:00:00.019”,”1970-01-01 08:00:00.02”,”1970-01-01 08:00:00.021”,”1970-01-01 08:00:00.022”],”val”:[20.0,21.0,22.0,23.0,24.0,25.0,26.0,27.0,28.0,29.0,30.0,31.0]},”schema”:”ts TIMESTAMP,val DOUBLE”} |

t