A couple of months ago, I was on a research project and had a problem to solve related to time series.
The problem was quite simple:
“From this time series with t time stepspredict the next k values“
For machine learning enthusiasts, this is like writing “Hello world”, as this problem is well known in the community by the name “forecast”.
The Machine Learning community developed many techniques that can be used to predict the next values of a time series. Some traditional methods involve algorithms like ARIMA/SARIMA either Fourier transform analysisand other more complex algorithms are Convolutional/recurrent neural networks or the super famous”Transformer”one (the T in ChatGPT stands for transformers).
While the problem of forecasting is well known, it is perhaps less unusual to address. with limitations.
Let me explain what I mean.
You have a time series with a set of parameters. x and the passage of time t.
He standard The weather forecasting problem is as follows:
The problem we face is the following:
So, if we consider that the input parameter has d dimensions, I want the function for dimension 1 (for example) to be monotonic. So how do we deal with that? How do we forecast a “monotonic” time series? The way we are going to describe this problem is XGBoost.
The structure of this blog post is as follows:
- About XGBoost: In a few lines we will describe what XGBoost is aboutWhat is the fundamental idea and what are the pros and cons.
