Quantitative transaction factor prediction method and device

Abstract

The invention provides a quantitative transaction factor prediction method and device, and the method comprises the steps: inputting obtained to-be-tested data into a pre-trained target GAN (Generative Adversarial Network) model, and obtaining a plurality of groups of simulation to-be-tested data outputted by the target GAN model; selecting three groups of to-be-tested data with low correlation from the plurality of groups of simulated to-be-tested data; and inputting the three groups of to-be-tested data into a pre-trained target TCN model to obtain a prediction result of the quantitative transaction factor corresponding to the to-be-tested data output by the target TCN model, so that the problems that the prediction of the quantitative transaction factor does not support parallelism and the prediction result is not accurate enough in the related technology can be solved, simulation data is generated by applying a GAN network, and the prediction accuracy of the quantitative transaction factor is improved. The sample size is increased, the diversity of samples is improved, and prediction is more reasonable; and parallel prediction of multiple variables and multiple time steps can be realized by using the TCN model, and the prediction accuracy is improved.

Description

technical field [0001] The present invention relates to the field of data processing, in particular to a method and device for predicting quantitative transaction factors. Background technique [0002] Monte Carlo methods are used to simulate the probability of different outcomes in processes that cannot be easily predicted due to the intervention of random variables, helping to account for the impact of risk and uncertainty in forecasting and forecasting models. The basis of Monte Carlo simulation involves assigning multiple values ​​to an uncertain variable to obtain multiple outcomes, and then averaging the results to obtain an estimate. Usually, more calculation steps and large samples are required for prediction and estimation to improve the accuracy, which requires relatively high calculation costs. It is necessary to assume a probability distribution, that is, random numbers must satisfy certain rules. Such a priori assumptions are not in line with the factual laws ...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a method and device for predicting quantitative transaction factors, so as to at least solve the problems in the related art that the prediction of quantitative transaction factors does not support parallelism and the prediction results are not accurate enough

Images