Forecast Data Trends with Confluent Platform for Apache Flink

Forecasting is an important use-case for Apache Flink®. You can deploy forecasting jobs using Confluent Manager for Apache Flink (CMF), a central management component of Confluent Platform for Apache Flink that enables users to securely manage a fleet of Flink jobs across multiple environments.

How does forecasting work?

Forecasting helps predict future values in time-series data by analyzing historical trends. In real-time streaming environments, continuous forecasting enables businesses to anticipate changes, optimize operations, and detect deviations as data flows in.

With Apache Flink®, you can perform continuous forecasting and anomaly detection directly within your familiar SQL-based environment.

Flink provides the Flink SQL forecast function, which enables you to perform real-time analysis and gain actionable insights from your streaming data without needing in-depth data science expertise. The ML_FORECAST function uses ARIMA (Autoregressive Integrated Moving Average), optimized for real-time performance, to deliver accurate forecasts in a streaming context.

Confluent Platform for Apache Flink also provides AI_FORECAST, which forecasts with pretrained time-series foundation models that run in model servers in your Kubernetes cluster. It needs no model parameters, forecasts from the 20th value of a series onward by default, and returns a range for each forecast step. For details, see Forecast with AI_FORECAST.

Your data must include:

• A timestamp column.

• A target column representing some quantity of interest at each timestamp.

ARIMA model

ARIMA, which stands for Autoregressive Integrated Moving Average, is a powerful statistical technique used for time series analysis and forecasting. It combines three key components to model and predict future values based on past data:

• Autoregressive (AR): This component uses past values of the time series to predict future values. It assumes that current observations are influenced by previous observations.

• Integrated (I): This refers to the differencing of raw observations to make the time series stationary. Stationary data has consistent statistical properties over time, which is crucial for accurate forecasting.

• Moving Average (MA): This component incorporates the relationship between an observation and a residual error from a moving average model applied to previous observations.

ARIMA models are used widely in various fields, including finance, economics, and weather forecasting, to analyze time-dependent data and make predictions. They are particularly effective when dealing with non-seasonal time series data that exhibits trends or patterns over time.

The model is typically represented as ARIMA(p,d,q), where:

• p: The number of autoregressive terms

• d: The degree of differencing

• q: The order of the moving average

By adjusting these parameters, analysts can fine-tune the model to best fit their specific time series data and improve forecasting accuracy.

Be aware that forecasting accuracy can vary greatly, based on many parameters, and there is no guarantee of correctness in predictions made by using ARIMA and machine learning.

Forecasting parameters

The following parameters are supported for forecasting.

Model configuration

Use the following parameters to configure the ML_FORECAST function. Pass them as a JSON string in the optional third argument, ML_FORECAST(value, timestamp[, config]), for example with JSON_OBJECT. If you omit the configuration argument, the function uses the defaults below, except that enableStl is TRUE.

enableStl

• Default: FALSE (TRUE if you omit the configuration argument)

Enable STL (Seasonal-Trend decomposition using Loess smoothing) decomposition. If set to TRUE, the time series is decomposed into trend, seasonal, and remainder components.

STL is a statistical method that separates a time series into three additive components: trend, seasonal, and remainder.

• The trend component captures the long-term direction or movement in the data.

• The seasonal component captures regular, predictable patterns that repeat over fixed periods.

• The remainder component captures random fluctuations that aren’t explained by trend or seasonality.

This decomposition helps improve forecasting accuracy by enabling the ARIMA model to better analyze and predict the underlying patterns in your data.

evalWindowSize

• Default: 5

• Validation: Must be >= 1 and <= 10000

Keeps the last evalWindowSize squared errors for both RMSE and AIC calculations. RMSE uses the rolling sum of these errors, while AIC uses the count (n) and sum of those same errors. A smaller window makes both metrics more responsive to recent errors; a larger window smooths out short-term noise.

An error is recorded only when a row arrives at the timestamp the model forecast, which is the previous row’s timestamp plus the most common interval between rows. If your timestamps are unevenly spaced, rmse and aic can stay NULL.

horizon

• Default: 5

Number of future time periods to forecast. This value defines how many steps ahead the model predicts.

m

• Default: If enableStl is TRUE, seasonality is estimated using ACF.

• Validation: m >= 0 and m <= (minTrainingSize / 2). Setting m requires enableStl to be TRUE.

The seasonal period length. If enableStl is TRUE, the model uses an ACF (Autocorrelation Function) to estimate the seasonal period length. The ACF measures the correlation between a time series and its lagged versions at different time intervals. This helps to identify repeating patterns and seasonal cycles in the data by detecting how strongly current values relate to past values at various lags.

maxTrainingSize

• Default: 128

• Validation: minTrainingSize <= maxTrainingSize <= 10000

Maximum number of historical data points used for training. This value limits the training window to recent data. The default equals the default minTrainingSize. If you set minTrainingSize above 128, also raise maxTrainingSize, or the configuration fails validation.

minTrainingSize

• Default: 128

• Validation: 1 <= minTrainingSize <= maxTrainingSize

Minimum number of data points required for training. The model won’t train until the quantity of available data is greater than this value.

p

• Default: NULL

• Validation: If provided, must be >= 0

Order of the AR (Autoregressive) term in the ARIMA model, which defines the number of lagged values used in the model. This value represents the number of past values that directly influence the current prediction.

If p is not provided, the model uses auto-ARIMA to determine the best order.

q

• Default: NULL

• Validation: If provided, must be >= 0

Order of the MA (Moving Average) term in the ARIMA model, which defines the number of lagged error terms in the model. This value represents the number of past prediction errors that influence the current prediction.

If q is not provided, the model uses auto-ARIMA to determine the best order.

d

• Default: NULL

• Validation: If provided, must be >= 0

Order of differencing in the ARIMA model, which defines the number of times the data is differenced to achieve stationarity. A value of 0 means no differencing, 1 means first difference, and so on.

If d is not provided, the model uses auto-ARIMA to determine the best order.

updateInterval

• Default: 1

• Validation: Must be > 0

Frequency of model updates. This value defines how often the model retrains with new data.

Forecast results

The ML_FORECAST function returns an array with one element for each of the horizon forecast steps. Each element is a row with the following fields, in the order timestamp, forecast_value, lower_bound, upper_bound, rmse, and aic. The array is empty until the model has trained.

aic

• Type: DOUBLE

The AIC (Akaike Information Criterion) at the time the forecast was generated, calculated over the last evalWindowSize errors.

AIC is a statistical measure used to evaluate and compare the quality of different models. It balances model accuracy against complexity by penalizing models with more parameters. Lower AIC values indicate better model performance, which helps to identify the optimal model configuration that provides the best fit to the data without overfitting.

forecast_value

• Type: DOUBLE

The forecast value at the timestamp.

lower_bound

• Type: DOUBLE

The lower confidence interval for the forecasted value.

rmse

• Type: DOUBLE

The RMSE (Root Mean Square Error) at the time the forecast was generated, calculated over the last evalWindowSize errors.

RMSE measures the difference between the predicted and actual values in a time series by calculating the square root of the average of the squared differences between the predicted and actual values. A lower RMSE indicates better model performance, because it means the model’s predictions are closer to the actual values.

timestamp

• Type: TIMESTAMP

The LocalDateTime for this forecast.

upper_bound

• Type: DOUBLE

The upper confidence interval for the forecasted value.

Example

1. Create a test table that has a timestamp field for your time series data.

   CREATE TABLE orders (
       total_orderunits double,
       summed_ts TIMESTAMP(3) NOT NULL,
       WATERMARK FOR summed_ts AS summed_ts
   );
   

2. Insert mock time series data into the table.

   INSERT INTO orders (total_orderunits, summed_ts) VALUES
   (102.12, TIMESTAMP '2024-11-19 10:00:00.000'),
   (103.45, TIMESTAMP '2024-11-19 10:01:00.000'),
   (101.89, TIMESTAMP '2024-11-19 10:02:00.000'),
   (104.23, TIMESTAMP '2024-11-19 10:03:00.000'),
   (102.78, TIMESTAMP '2024-11-19 10:04:00.000'),
   (102.12, TIMESTAMP '2024-11-19 10:05:00.000'),
   (103.45, TIMESTAMP '2024-11-19 10:06:00.000'),
   (101.89, TIMESTAMP '2024-11-19 10:07:00.000'),
   (104.23, TIMESTAMP '2024-11-19 10:08:00.000'),
   (102.78, TIMESTAMP '2024-11-19 10:09:00.000'),
   (102.12, TIMESTAMP '2024-11-19 10:10:00.000'),
   (103.45, TIMESTAMP '2024-11-19 10:11:00.000'),
   (101.89, TIMESTAMP '2024-11-19 10:12:00.000'),
   (104.23, TIMESTAMP '2024-11-19 10:13:00.000'),
   (102.78, TIMESTAMP '2024-11-19 10:14:00.000'),
   (102.12, TIMESTAMP '2024-11-19 10:15:00.000'),
   (103.45, TIMESTAMP '2024-11-19 10:16:00.000'),
   (101.89, TIMESTAMP '2024-11-19 10:17:00.000'),
   (104.23, TIMESTAMP '2024-11-19 10:18:00.000'),
   (102.78, TIMESTAMP '2024-11-19 10:19:00.000'),
   (102.12, TIMESTAMP '2024-11-19 10:20:00.000'),
   (103.45, TIMESTAMP '2024-11-19 10:21:00.000'),
   (101.89, TIMESTAMP '2024-11-19 10:22:00.000'),
   (104.23, TIMESTAMP '2024-11-19 10:23:00.000'),
   (102.78, TIMESTAMP '2024-11-19 10:24:00.000'),
   (102.12, TIMESTAMP '2024-11-19 10:25:00.000'),
   (103.45, TIMESTAMP '2024-11-19 10:26:00.000'),
   (101.89, TIMESTAMP '2024-11-19 10:27:00.000'),
   (104.23, TIMESTAMP '2024-11-19 10:28:00.000'),
   (102.78, TIMESTAMP '2024-11-19 10:29:00.000');
   

1. Run the following statement to perform a forecast on the mock data in the orders table.

   SELECT
       ML_FORECAST(total_orderunits, summed_ts, JSON_OBJECT('p' VALUE 1, 'q' VALUE 1, 'd' VALUE 1, 'minTrainingSize' VALUE 10, 'enableStl' VALUE false, 'horizon' VALUE 5))
       OVER (
           ORDER BY summed_ts
           RANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) AS forecast
   FROM orders;
   

2. Run the following statement to perform a forecast with AutoArima on the mock data in the orders table.

   SELECT
       ML_FORECAST(total_orderunits, summed_ts, JSON_OBJECT('minTrainingSize' VALUE 10, 'enableStl' VALUE false))
       OVER (
           ORDER BY summed_ts
           RANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) AS forecast
   FROM orders;
   

Forecast with AI_FORECAST

Important

AI_FORECAST is a preview feature. Interfaces might change in future releases.

AI_FORECAST sends the recent values of a series to a time-series foundation model, and returns the model’s forecast of the next values, each with a range. The models are pretrained, so the function needs no parameters and no training data. With the default settings, it forecasts from the 20th value of a series onward.

The models run in model servers in your Kubernetes cluster. Before you can call the function, an administrator must deploy the model servers and register the function in your environment, as described in Set Up AI Functions in Confluent Manager for Apache Flink.

Syntax

AI_FORECAST(value, timestamp[, config])
  OVER (ORDER BY timestamp RANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW)

value

The value of the series, as a DOUBLE. The function leaves rows whose value is NULL, NaN, or infinite out of the series.

timestamp

The time of the value, as a TIMESTAMP(3). The function takes the series’ step from the median interval between timestamps, and gives each forecast step the timestamp of the last value plus a multiple of that step, so the timestamps of a series should be evenly spaced.

config

Optional. A JSON object, as a string or from JSON_OBJECT, with the settings in the following table. The configuration must be the same for every row of a series.

Call the function from a statement whose current catalog is the environment catalog, _env_<environment>, and whose current database is default. To forecast several series in one statement, add PARTITION BY to the OVER clause.

Configuration

Key	Default	Description
model	ttm	The model to forecast with: ttm, flowstate, or patchtst. The environment must have a model server for it.
horizon	5	The number of steps to forecast, at least 1. The model servers refuse more than 96.
minContextSize	20	The number of values the function waits for before it forecasts. At least 2.
maxContextSize	512	The largest number of recent values that the function sends to the model. At least minContextSize.

Keys are case-insensitive. For example, the following call forecasts 12 steps with FlowState:

AI_FORECAST(v, ts, JSON_OBJECT('model' VALUE 'flowstate', 'horizon' VALUE 12))

Results

AI_FORECAST returns a row with the following fields:

Field	Type	Description
forecast	ARRAY<ROW<timestamp TIMESTAMP(3), mean DOUBLE, q10 DOUBLE, q50 DOUBLE, q90 DOUBLE>>	One element for each forecast step: its timestamp, the forecast value, and the 10th, 50th, and 90th percentile of the forecast. The array is empty until the series has minContextSize values.
metadata	STRING	A JSON object from the model server that describes the model that made the forecast.

Arrays in Flink SQL start at index 1, so forecast[1] is the first step. timestamp is a reserved word, so quote it with backticks when you read the field, as the following example does.

Example

1. Create a table that generates a series with one value per second, in the environment catalog:

   CREATE TABLE sensor_readings (
     id BIGINT,
     ts AS CAST(TIMESTAMPADD(SECOND, CAST(id AS INT), TIMESTAMP '2026-01-01 00:00:00') AS TIMESTAMP(3)),
     WATERMARK FOR ts AS ts
   ) WITH (
     'connector' = 'datagen',
     'fields.id.kind' = 'sequence',
     'fields.id.start' = '1',
     'fields.id.end' = '10000',
     'rows-per-second' = '20'
   );
   

2. Forecast the next five values of a wave with a period of 24 seconds, and read the first step:

   SELECT
     id,
     v,
     CARDINALITY(f.forecast) AS steps,
     f.forecast[1].`timestamp` AS next_ts,
     f.forecast[1].mean AS next_mean,
     f.forecast[1].q10 AS next_q10,
     f.forecast[1].q90 AS next_q90
   FROM (
     SELECT
       id,
       v,
       AI_FORECAST(v, ts, JSON_OBJECT('model' VALUE 'ttm'))
         OVER (ORDER BY ts RANGE BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) AS f
     FROM (
       SELECT id, ts, CAST(10 + 5 * SIN(2 * PI() * id / 24) + 0.3 * SIN(id * 1.7) AS DOUBLE) AS v
       FROM sensor_readings
     )
   );
   

   The first 19 rows have no steps. From row 20 on, each row has five steps, and next_ts is one second after the row’s timestamp.

Limitations

• Each row sends one request to the model server and waits for the answer, so a statement processes rows only as fast as its model server answers.

• timesfm-2.5 is not available, because no model server for it is provided.

For the errors that the function reports, see the troubleshooting section of Set Up AI Functions in Confluent Manager for Apache Flink.

Related content

• Set Up AI Functions in Confluent Manager for Apache Flink

• Anomaly Detection with Confluent Manager for Apache Flink

• Submit Flink SQL Statement with Confluent Manager for Apache Flink

• Deploy and Manage Statements in Confluent Manager for Apache Flink

Note

This website includes content developed at the Apache Software Foundation under the terms of the Apache License v2.