Simsets API

My new web API, simsets, generates simulated data for some common scenarios. Why is this useful? Because if you generated the data then you know the answers to many questions that are unanswerable in real life. Some examples: What is the true contribution of advertising to a time series of product sales? Which latent factors really explain this viewing data? With real life data, modelling allows you to estimate such things but you can't know that you are right.

So having the details of the simulation is like having the answer sheet at the back of the book. This makes simulated perfect for use-cases like:

• Testing forecasting and prediction methods
• Testing methods which try to explain the data using other observable or latent variables
• Testing the claims of service providers who say they can do either of the above
• Creating interview questions
• Creating examples for teaching

The simsets API saves you the trouble of producing this data. It gives you the data and the answers.

The first endpoint simulates explainable time series data. Try it here. And try refreshing it. Note the data set is downloadable from the webpage. Note also the model is fully described.

And here is another endpoint that generates viewing data for an imaginary video on demand service (Simflix).

In both cases to access the data programatically use output_type=json as in this example. Note the json includes the latex for model.

The code is available on github, and there are examples of using the API in Jupyter for the explainable time series and the Simflix viewing data.

Hope it is useful!