This is an open-access article distributed under the terms of the creative Commons Attribution License ( CC BY ). The use, distribution or reproduction in other forums is permitted, provided the original writer ( second ) and the copyright owner ( s ) are credited and that the original publication in this journal is cited, in accordance with accept academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. The health insurance diligence in China is undergoing big shocks and profound impacts induced by the cosmopolitan COVID-19 pandemic. Taking for example the three prevailing listed companies, namely, China Life Insurance, Ping An insurance, and Pacific Insurance, this composition investigates the equity performances of China ‘s health policy companies during the pandemic. We first construct a stock monetary value forecasting methodology using the autoregressive desegregate moving average, bet on generation neural network, and long short-run memory ( LSTM ) nervous network models. We then empirically study the stock monetary value performances of the three listed companies and find out that the LSTM model does better than the other two based on the criteria of base absolute error and intend feather error. ultimately, the above-mentioned models are used to predict the stock price performances of the three companies .


The COVID-19 pandemic is first and foremost induce imbalance and uncertainty in the high-quality development of China ‘s economy, not only impacting industries of exile and tourism but besides causing significant fluctuations in the fiscal commercialize ( 1 – 4 ). As a barometer of the economy, livestock prices reflect the current conditions and future trends in the exploitation of industries. During the pandemic, China ‘s breed market is on the whole refuse, while the stock prices of the health policy sector continue to rise as grocery store demand expands. It is of practical significance to study the standard price performances of China ‘s health policy companies during the pandemic and to provide insights into the impacts of the pandemic and trends in the development of this industry.

The impacts of the COVID-19 pandemic on the stock marketplace have become a focus in recent research. Duan ( 5 ) explored the impacts of the pandemic on the stock returns of list companies in China ‘s pharmaceutical diligence using the event analysis method acting. The empiric results show that the pandemic is having a significantly incontrovertible short-run impact on the stock returns in this industry. Based on data of 3,550 listed companies in China in the early degree of the pandemic, Wang et alabama. ( 6 ) studied the impacts of new confirmed cases in the place of company registration on the fluctuations of neckcloth prices. Using the panel vector autoregression model, individual fixed effects exemplary, and moral force econometric model, they found a u-shaped relationship between daily stock prices and numbers of modern confirmed cases. similarly, Xia and Hu ( 7 ) used the Fama-French three-factor mannequin and control panel regression to analyze the performances of 223 pharmaceutical stocks during the pandemic in Shanghai and Shenzhen 300 exponent. Sun et aluminum. ( 8 ) found out, in their character studies, about a stronger plus correlation coefficient between investor opinion and stock certificate returns during the pandemic than in previous periods. Mazur et alabama. ( 9 ) explored the stock market performances in the United States during the pandemic and found that the stock prices in natural boast, food, healthcare, and software industries showed an up drift, while the stock prices in vegetable oil, substantial estate of the realm, entertainment, and hotel industries showed the opposition. Heyden and Heyden ( 10 ) studied the short-run reaction in stock markets of the United States and Europe at the early stage of the pandemic. Their results showed that fiscal measures had a negative impact on stock returns, while monetary policy had a brace effect on the market.

Stock price fluctuations during the pandemic directly affect the constancy of the fiscal marketplace and healthy growth of national economy. thus, the prediction of stock price performances becomes a hot topic in holocene research. The most often use methods include the autoregressive integrate moving average ( ARIMA ), back propagation ( BP ) neural network, and long short-run memory ( LSTM ) neural network models. Bai ( 11 ) and Shi et alabama. ( 12 ) used the ARIMA method acting to model the stock prices of the Shanghai composite exponent, established an better model in their short-run prediction, and then proved the potency of both models. Chen ( 13 ) constructed the ARIMA and BP nervous net models to predict the broth prices of two celebrated companies in the IT industry of China, namely, Baidu and Alibaba. Both models are found to have ideal prediction accuracy and short-run prediction effect. Some research specifically aim to tackle the excess problem of the experimental samples. For example, Cai and Chen ( 14 ) and Huo et aluminum. ( 15 ), respectively, proposed the broth price prediction models of the principal component analysis–BP nervous network and LM–BP algorithm and verified their high accuracy in short-run prediction. With a view to improve prediction accuracy, Peng et aluminum. ( 16 ) constructed the LSTM nervous network mannequin of different layers for lineage price prediction and found out about the allow numbers of LSTM layers and hide neurons. furthermore, Song et aluminum. ( 17 ) proposed a LSTM nervous network model based on particle drove optimization, which matched the characteristics of malcolm stock prices with network topology thus as to improve prediction accuracy. however, inquiry and studies have largely focused on the sprout price performances in industries of real estate, Internet, medicine, and some others. With the COVID-19 pandemic ongoing and health insurance becoming a most significant foundation of people ‘s support, the prediction of broth prices in this industry is of great meaning for analyzing its prospects. Seeing this, this newspaper constructs a livestock price prediction method acting using the ARIMA, BP nervous network, and LSTM nervous network models to study the impacts of the pandemic on China ‘s health policy industry from the perspective of lineage price fluctuations. This paper is organized as follows : section Trend Analysis of Stock Closing Prices makes a descriptive statistical analysis of the stock prices of the three dominant health insurance listed companies during trading days from 2015 to 2020. In section Methodology, we introduce three prediction models, namely, the ARIMA, BP nervous net, and LSTM nervous network models. Those models are applied to the malcolm stock price predictions of the three companies in section Stock Price Prediction, and a relative analysis of the empiric results using respective models is besides given. section Conclusion and Prospect gives the conclusion of this newspaper .

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: