Dark Mode
Synthetic Retail Forecasting Data
Synthetic Data Generation
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Synthetic sales data specifically engineered with time-series features. It is ideally suited for use in testing and exploring various forecasting models. The data has been tuned to reflect weekly seasonality, with a seasonality coefficient set to 7, meaning weekends consistently portray higher sales figures than weekdays. The data is entirely scalable, allowing users to increase the record count simply by adjusting the parameters of the accompanying generation code.
Columns
The dataset contains three distinct columns:
- Sr. No.: A sequential serial number or index for each record, ranging from 0 up to 364.
- Date: The daily timestamp, formatted as Date/DateTime. The range spans exactly one year.
- Sales: The synthetic daily sales figure, which varies between 19 and 48.2, with a mean of 34.1.
Distribution
The dataset is available as a single CSV file named
sales.csv, which is approximately 12.45 kB in size. It currently holds 365 valid records. The data is exceptionally clean, with 100% validity and no missing or mismatched data points across all three fields.Usage
Ideal applications include:
- Forecasting Model Validation: Use this clean dataset to rigorously test and benchmark different time-series forecasting algorithms.
- Seasonality Study: Explore how models perform when explicitly faced with defined weekly seasonality, where sales spikes occur on weekends.
- Educational Projects: Provide reliable, synthetic data for academic or learning exercises focused on time-series analysis.
- System Stress Testing: Leverage the scalability of the data (by tweaking parameters) to create larger datasets for stress-testing data pipelines.
Coverage
The temporal scope of the records covers a full year, specifically running from 1 October 2021 through to 30 September 2022. As the data is synthetic and generated programmatically, there is no associated geographic or demographic scope.
License
CC0: Public Domain
Who Can Use It
- Data Scientists: For training and evaluating predictive models, particularly those involved in sales or inventory forecasting.
- Machine Learning Engineers: To efficiently generate repeatable, clean test data for pipeline development and quality assurance.
- Academics and Students: Individuals learning time-series analysis who require a guaranteed high-quality, non-proprietary dataset.
Dataset Name Suggestions
- Synthetic Retail Forecasting Data
- Time-series Sales Model Test Set
- Scalable Sales Forecasting Data
- Weekly Seasonality Sales Data
Attributes
Original Data Source: Synthetic Retail Forecasting Data
QUALITYLoading...