Abstract
The article proposes an innovative approach to longterm forecasting based on a multitrend forecast. The approach is based on the decomposition of a time series into separate components – trends, which allows the model to take into account such important components of dynamics as the general upward/downward trend (logarithmic or linear trend), cyclicity (sinusoidal trend). The initial data for the analysis is a time series, for example, the prices of the company's shares. Before applying the developed forecast model, the data is preanalyzed in order to identify the stationary sites. Further, the trend components are consistently highlighted. First, the parameters of the logarithmic function are selected. Then the found dependence is subtracted from the series and a similar procedure is applied to the residuals to determine the coefficients of the linear trend. After that, sinusoidal trends of various periods describing cyclic fluctuations are sequentially subtracted from the series. The stopping criterion is the achievement of a white noise state time series when there are no trends in the series anymore. Then all the resulting component trends are summarized to build a final forecast for the future period. The algorithm described above was implemented in the MS Excel spreadsheet environment.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
