Decision Tree Number Space - Simple

The number space or number universe is important to consider when developing models but especially when developing decision trees. There are two important features of data (measure theory) that need to be considered. 1) Density 2) Range and Domain Density impacts the reliability of the leaf estimates. When estimating a leaf by using the mean, there needs to be enough data to estimate the mean. Range and Domain have to do with the total number of values covered within the number space. Models are always built off of samples and not the population. It can be challenging to determine the appropriateness of the sample as the population is never fully known. When areas of the population on never used in the estimation of a decision tree, it will fail depending how different those unknown areas are. In quantitative finance, moving into new products or credit ranges can make decision trees invalid. Website: https://www.FancyQuantNation.com Support: https://ko-fi.com/fancyquant Quant t-shirts, mugs, and hoodies: https://www.teespring.com/stores/fancy-quant Connect with me: https://www.linkedin.com/in/dimitri-bianco https://twitter.com/DimitriBianco