How to Find Similar Neighborhoods in a Time Series?

The aim of the paper is to turn the models parometer. So we have the dalgorism, try to tosik among some values of neighbors and fister selection the best parometers of nearest neighbors. And this by implement a method inspired by cross validation techniques. To forecast the samptaries by nearest neighbors, we need to detemine wis, the volues of its barometers,. Thas are the number of neighbors, and fisser selection. A small alteration of canaries neighbors barometers can reduce the models performance.