Comparison of models with neural network and OLS-regression in constructing the risk management strategy against the income according to index

Introduction. The models with neural network and OLS-regressions are used in the stock market and include variables that describe the state of the stock market. One of the possible ways to determine these dependencies is clusterization trough analizing principal components. The main aim of the research is revealing the essence of two promising heuristic approaches to assessment of the dynamics of functional relationships between the incomes in the stock market and variables that describe the state of the market. Materials and Methods. The source data are models with a continuous network and OLS-regression in the area of management strategies. Mathematical statistics revenue management strategies. Results. It is well known that specifics of functional relationship establishment between the income in the stock market lies in their clusterization through a linear (nonlinear) analysis of principal components of the market condition. We analyzed two promising heuristic approaches to the assessment of the dynamics of functional relationships between the income in the stock market and variables describing the state of the market. Discussion and Conclusions. The analysis of the dynamics of functional links between the revenues on the stock market was made.