Research and Optimazation on Evaluation Index Prediction Using Grey Neural Network and Big Data

Aiming at the research problem of the valuation level of the stock market and its quantitative relationship with the fundamental index and liquidity index, this paper makes all the original data fall into the tolerable coverage domain through the method of series level ratio. The random error term of time series is eliminated by summation of sequence in grey prediction model. The cumulative time series is used as the input layer neuron of the neural network, and the grey neural network prediction model is established to quantitatively describe and predict the valuation level, and the weighted average price/sales ratio is used to obtain the overall valuation level of the science and technology innovation.