Methods

In forecasting, the key is not the method but the fit between data and the task. Every forecast depends on which data we use and what remains after removing noise. There are three levels. Raw data. Observations, prices, news. Without interpretation — just a stream. Processing. Smoothing, filters, feature selection. An attempt to preserve structure and remove the rest. Models. From simple heuristics to statistics and ML. The model should be simple enough to work, and complex enough to reflect probability. ...

In forecasting, there are three types of models. Models with predictors. They use external variables — temperature, the economy, human behavior. These models explain why something happens. Time‑series models. They look only at the past values of the series. They explain nothing, but often predict better. Hybrid models. A combination of past dynamics and external factors. Why is it sometimes better to use only a time series? Because external variables may be unknown, hard to measure, or unpredictable themselves. Sometimes the goal is simply to predict, not to explain. ...