Sales Forecasting Model Using SARIMAX Algorithm

Tasks

• This bot can handle data with a trend and explicitly supports univariate time series analysis with a seasonal component.
• The bot takes in the CSV file as input and performs the Seasonal ARIMAX and provides output in Excel.
• Help in performing seasonal time series analysis for a marketer to make data driven business decisions.

This bot forecasts trends such as seasonality for traffic, purchases, or other data you need to plan your business.

Use Cases:

• An organization looking to forecast Daily, Weekly, Monthly product sales while being cautious of seasonality.
• A retail store planning inventory levels based on sales forecasts and seasonal variations.

Inputs:

• Path: Path of the CSV file. Ex.: C:~sales data-set.csv
• Column Count: Count of columns available in CSV file being provided with input data
• Input Column Names: Name of columns being provided in CSV file. For example 'Month', 'Sales'
• Input Column Data Type: Data Type of columns being provided in CSV file in the same order as column names. For example 'string', 'Date', 'Numeric'
• Missing Number Value Replacement: A value that can be used by the model in case CSV file has blank fields or missing numerical values
• Missing Date Value Replacement: A value that can be used by the model in case the CSV file has a blank field or missing a date value. This parameter should have value in the YYYY-MM-DD format. Ex: 2020-01-01
• Ignore Columns: Names of the column that model shall ignore while estimating forecasting. For example 'Sno'

Input Sample:

• Retail Sales data of 105 months (1964-01-01 to 1972-09-01)

Actions:

• The bot takes a CSV file as an input, performs the time-series analysis using SARIMAX algorithm and provides an output in Excel to forecast sales.

Outputs:

• Excel file
• Seasonal time series with predictions including upper and lower limits