Time Series ML Forecasting#

Description:

Machine Learning Time Series model to predict future values. This can be based on daily, monthly, or hourly values.

Function:

TS_Forecast(table = string, date_col = string, value_col = string, num_per = integer, freq = string, new_table_name = string)

Parameters:

  • Table: Table name on which to perform the function

  • Date Column: Date column to use for forecast

  • Value Column: Value Column to base forecast on

  • Number of Periods: Number of Periods to Forecast

  • Forecast Frequency: What is the time interval to be forecasted (hourly, daily, monthly)

  • New Table Name: Name for the new table

Example:

TS_Forecast(table = Budget, date_col = "Date", value_col = "Revenue", num_per = 5, freq = "Daily", new_table_name = "TS Forecasting")
Before#

Date

Value

6/26/2024

1178

6/27/2024

1085

6/28/2024

935

6/29/2024

1342

6/30/2024

1087

After#

Date

FORECAST

Predicted Value 1

Predicted Value 2

Predicted Value 3

2024-06-29T00:00:00.000000

1001.975

751.9004

1256.553

1001.975

2024-06-30T00:00:00.000000

1004.383

761.6355

1233.621

1004.383

2024-07-01T00:00:00.000000

1001.932

761.8277

1266.305

1001.932

2024-07-02T00:00:00.000000

1004.306

768.8828

1272.184

1004.306

2024-07-03T00:00:00.000000

1000.19

754.8353

1259.737

1000.19

2024-07-04T00:00:00.000000

1131.743

898.6075

1379.402

1131.743

2024-07-05T00:00:00.000000

1004.984

772.052

1255.279

1004.984