Model for Assessing the Liquidity of a Stock Market Trading Instrument

Currently, a large number of studies are being conducted to improve the accuracy of the developed forecasting methods for the stock market. At the same time, multivariate models based on machine learning methods are increasingly used. Since liquidity indicators have a significant impact on asset pricing, taking them into account can improve the accuracy of forecasting. The purpose of this study is to develop machine learning models that forecast securities quotes taking into account the liquidity factor, as well as to analyze the impact of liquidity on the accuracy of forecasting various types of securities. Using the example of multivariate models ARIMA and LSTM, a study was conducted of forecast indicators of stock quotes with the addition of a feature time series with liquidity ratios. The results of the study show that taking liquidity into account is of great importance in developing more accurate forecasting methods, which is very important for investors and investment companies.