Forecasting of the balance sheet with a non-zero trust level on the basis of univariate linear regression

The subject of research is financial forecasting. The work has as its objective the development stated in the subject of the approach using interval forecasts of revenues from product sales. The study used tools of mathematical statistics and probability theory. An approach to the design of the balance sheet is proposed, based on a three-level method for justifying reliability and a two-level method for interval forecasting. The method of substantiating reliability involves testing the hypothesis of a nonrandom relationship between: 1) possible in time estimates of revenues from the sale of products; 2) possible in time factors of the dynamics of revenue from the sale of products; 3) possible estimates of the proceeds from the sale of products and possible estimates of the balance sheet item. The method of interval forecasting assumes a mathematically formalized justification: 1) proceeds from the sale of products through extrapolation; 2) the state of the balance sheet through a single-factor linear regression. The presented approach can be used by enterprises of the real sector of the economy (where the main form of income is the revenue from sales) to estimate the necessary size of the economic tools and the appropriate level of funding sources, without which it is difficult to obtain the forecast value of revenue from sales of products. The developed approach makes it possible to justify the level of confidence and interval forecasting of the state of the balance sheet.