摘要：Heat supply is socially and economically important in our
country. In this regard, high-quality monitoring and planning of the
development of heat supply systems are a strategic vector of scientific
research. This paper is focused on the studies demonstrating how to choose
a methodological approach to describe changes in heat consumption in the
retrospective. The change in heat consumption is described using multiple
regression models. In the first part of the paper, the parameters for the
regression model are determined and a statistical analysis of the obtained
model is performed. In the second part of the paper, to eliminate the
multicollinearity of the regression equation, the number of dependent
variables in the model is reduced. A statistical analysis of the new
regression model and the exponential regression model are carried out. The
heat consumption values obtained using these models are compared with
the statistical data. The conclusions about the quality of the obtained
regression models are made. In the third part of the article, we make a
forecast of heat consumption in the medium term by using a linear
regression model and an exponential model.
其他摘要：Heat supply is socially and economically important in our country. In this regard, high-quality monitoring and planning of the development of heat supply systems are a strategic vector of scientific research. This paper is focused on the studies demonstrating how to choose a methodological approach to describe changes in heat consumption in the retrospective. The change in heat consumption is described using multiple regression models. In the first part of the paper, the parameters for the regression model are determined and a statistical analysis of the obtained model is performed. In the second part of the paper, to eliminate the multicollinearity of the regression equation, the number of dependent variables in the model is reduced. A statistical analysis of the new regression model and the exponential regression model are carried out. The heat consumption values obtained using these models are compared with the statistical data. The conclusions about the quality of the obtained regression models are made. In the third part of the article, we make a forecast of heat consumption in the medium term by using a linear regression model and an exponential model.
