AI Blocks › Track your numbers
Regression and forecast modeller
Give it a data table. Get a fitted regression or time series model with checks and a plain read of the results.
Regression and forecast modeller is a free AI skill for the job "Track your numbers". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · MIT licence · Made by K-Dense · Level: advanced
Made by K-Dense. See the original on GitHub; we copied it as it is, with its licence.
What it does
A fitted model for your data table, with the checks that show whether you can trust it, and a plain reading of what the numbers mean. It uses Python and the statsmodels library. You describe the question, such as what drives sales or what next quarter looks like. It picks a fit: ordinary regression, logistic or count models, mixed models, or time series like ARIMA. Then it tests the assumptions, looks for outliers and compares candidate models. A good fit only proves the maths ran, so it separates correlation from cause and warns you about both. It needs Python and a few packages installed.
You give it
- Data table as a CSV
- What you want to explain or predict
- Time column for forecasts (optional)
- Python set up with statsmodels
You get
- Fitted model with coefficients
- Assumption and outlier checks
- Forecast with ranges (optional)
- Plain language reading of results
Use it when
- You want to know which factors really move your sales or conversion numbers.
- You need a forecast from monthly figures with a clear range of uncertainty.
- A model gave odd results and you need to find out why.
- You must compare two models and choose one.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run code.