Deformed steel bar futures price rise and fall probability prediction method

Abstract

The invention discloses a deformed steel bar futures price rise and fall probability prediction method. The method comprises the following specific steps: collecting deformed steel bar characteristicdata from the Internet and a third-party database; The information gain ratio is combined with a square error minimization criterion, the characteristic of large information gain is reserved, and a decision tree is generated; Calculating the empirical entropy of each node through a loss function, recursively retracting upwards from leaf nodes of the tree, and if all leaf nodes of a certain fathernode are combined to reduce the loss function of the father node, pruning and changing the father node into a new leaf node; Repeating the steps until the combination cannot be continued, and finallyreducing the over-fitting probability; According to the method, the threaded steel futures price prediction speed is increased, the manual analysis cost is saved, multi-dimensional big data statistical analysis difficult to complete manually is achieved, meanwhile, the model has the characteristic of continuous learning, and the prediction precision can be higher and higher.

Description

technical field [0001] The invention relates to the field of futures price prediction, in particular to a method for predicting the probability of rise and fall of rebar futures prices. Background technique [0002] Definition of Terms: [0003] Rebar: Rebar is a common name for hot-rolled ribbed steel bars. The grade of ordinary hot-rolled steel bar is composed of HRB and the minimum yield point of the grade. H, R, and B are the English initials of the three words Hotrolled, Ribbed, and Bars, respectively. [0004] Futures: Futures (Futures) are completely different from spot goods. Spot goods (commodities) are actually tradable goods (commodities). Futures are not mainly goods, but are based on certain mass products such as cotton, soybeans, oil, etc. and financial assets such as stocks and bonds. Etc. is the target standardized tradable contract. Therefore, the subject matter can be a certain commodity (such as gold, crude oil, agricultural products), or it can be a f...

AI Technical Summary

Problems solved by technology

[0013] 1. It does not analyze the characteristics of rebar varieties and cannot be used to predict the rise and fall of rebar futures

[0014] 2. Does not use machine learning technology, does not have self-learning features, and needs to manually adjust parameters repeatedly, which is time-consuming and laborious

[0015] 3. The analysis speed is slow, it is not suitable for big data scenarios, and the workload of adding features is heavy

Images