Problems and prospects of using vector autoregressions in short-term forecasting of stock market conditions
Jazyk: | ruština |
---|---|
Rok vydání: | 2022 |
Předmět: | |
DOI: | 10.18720/spbpu/3/2023/vr/vr23-713 |
Popis: | Тема вÑпÑÑкной квалиÑикаÑионной ÑабоÑÑ Ð¼Ð°Ð³Ð¸ÑÑÑа: «ÐÑÐ¾Ð±Ð»ÐµÐ¼Ñ Ð¸ пеÑÑпекÑÐ¸Ð²Ñ Ð¸ÑполÑÐ·Ð¾Ð²Ð°Ð½Ð¸Ñ Ð²ÐµÐºÑоÑнÑÑ Ð°Ð²ÑоÑегÑеÑÑий в кÑаÑкоÑÑоÑном пÑогнозиÑовании конÑÑнкÑÑÑÑ ÑондовÑÑ ÑÑнков».ЦелÑÑ Ð¸ÑÑÐ»ÐµÐ´Ð¾Ð²Ð°Ð½Ð¸Ñ ÑвлÑеÑÑÑ ÑазÑабоÑка модели векÑоÑной авÑоÑегÑеÑÑии Ð´Ð»Ñ ÐºÑаÑкоÑÑоÑного ÑкономиÑеÑкого пÑогнозиÑÐ¾Ð²Ð°Ð½Ð¸Ñ ÐºÐ¾Ð½ÑÑнкÑÑÑÑ Ñондового ÑÑнка на пÑимеÑе оÑÑаÑлевÑÑ Ð¸Ð½Ð´ÐµÐºÑов «ÐоÑковÑкой биÑжи».ÐеÑÐ¾Ð´Ñ Ð¸ÑÑледованиÑ: поиÑк и ÑиÑÑемаÑизаÑÐ¸Ñ Ð¸Ð½ÑоÑмаÑии, маÑемаÑиÑеÑкое моделиÑование, анализ даннÑÑ , маÑемаÑиÑеÑкие вÑÑиÑлениÑ, ÑÑавниÑелÑнÑй анализ.ÐÑновнÑе ÑезÑлÑÑаÑÑ Ð¸ÑÑледованиÑ:-      изÑÑÐµÐ½Ñ ÐºÐ¾Ð½ÑепÑии иÑÑÐ»ÐµÐ´Ð¾Ð²Ð°Ð½Ð¸Ñ Ñондового ÑÑнка и изÑÑÐµÐ½Ñ Ð¼ÐµÑÐ¾Ð´Ñ ÐµÐ³Ð¾ анализа;-      ÑаÑÑмоÑÑена ÑволÑÑÐ¸Ñ ÐºÑаÑкоÑÑоÑнÑÑ Ð¼ÐµÑодов пÑогнозиÑованиÑ;-      ÑоÑÑÐ°Ð²Ð»ÐµÐ½Ñ Ð²ÑбоÑки набоÑов даннÑÑ Ð´Ð»Ñ Ð°Ð½Ð°Ð»Ð¸Ð·Ð° и поÑÑÑÐ¾ÐµÐ½Ñ Ð¼Ð¾Ð´ÐµÐ»Ð¸ векÑоÑной и комплекÑнознаÑной векÑоÑной авÑоÑегÑеÑÑии;-      вÑполнен ÑÑавниÑелÑнÑй анализ Ñ Ð°ÑакÑеÑиÑÑик моделей и вÑбÑана наиболее ÑÑÑекÑÐ¸Ð²Ð½Ð°Ñ Ñ ÑоÑки зÑÐµÐ½Ð¸Ñ ÑоÑноÑÑи пÑогнозиÑÐ¾Ð²Ð°Ð½Ð¸Ñ Ð¼Ð¾Ð´ÐµÐ»Ñ;-      пÑоведен анализ ÑоÑÑавленного пÑогноза.ÐблаÑÑÑ Ð¿ÑÐ¸Ð¼ÐµÐ½ÐµÐ½Ð¸Ñ ÑезÑлÑÑаÑов ÐÐÐ â кÑаÑкоÑÑоÑное ÑкономиÑеÑкое пÑогнозиÑование взаимозавиÑимÑÑ Ð¿Ð°ÑамеÑÑов Ð´Ð»Ñ ÑеÑÐµÐ½Ð¸Ñ ÑазлиÑнÑÑ Ð·Ð°Ð´Ð°Ñ ÑоÑиалÑно-ÑкономиÑеÑкого пÑоÑилÑ.ÐаÑÑÐ½Ð°Ñ Ð½Ð¾Ð²Ð¸Ð·Ð½Ð° иÑÑледованиÑзаклÑÑаеÑÑÑ Ð² Ñом, ÑÑо пÑименение комплекÑнознаÑнÑÑ Ð°Ð²ÑоÑегÑеÑÑий оÑкÑÑÐ²Ð°ÐµÑ Ð¿ÐµÑед пÑакÑикÑÑÑими аналиÑиками ÑиÑокие возможноÑÑи, ÑÐ½Ð¸Ð¶Ð°Ñ Ð²ÑÑиÑлиÑелÑнÑÑ ÑложноÑÑÑ Ð¾Ñенки коÑÑÑиÑиенÑов модели.ÐÑводÑ. Ð¦ÐµÐ»Ñ ÐÐРдоÑÑигнÑÑа â опÑеделена ÑÑÑекÑÐ¸Ð²Ð½Ð°Ñ Ð¼Ð¾Ð´ÐµÐ»Ñ ÐºÑаÑкоÑÑоÑного пÑогнозиÑованиÑ, поÑÑавленнÑе задаÑи в пÑоÑеÑÑе вÑÐ¿Ð¾Ð»Ð½ÐµÐ½Ð¸Ñ ÑеÑенÑ. ÐбознаÑена облаÑÑÑ Ð¿ÑÐ¸Ð¼ÐµÐ½ÐµÐ½Ð¸Ñ ÑезÑлÑÑаÑов ÐÐРи наÑÑÐ½Ð°Ñ Ð·Ð½Ð°ÑимоÑÑÑ Ð¿Ð¾Ð»ÑÑеннÑÑ Ð¸Ñогов. The topic of the graduate qualification work of the master's degree: "Problems and prospects of using vector autoregressions in short-term forecasting of stock market conditions".The purpose of the study is to create a model of vector autoregression for short-term economic forecasting of stock market conditions on the example of industry indices of the Moscow Exchange.Research methods: search and systematization of information, mathematical modeling, data analysis, mathematical calculations, comparative analysis.Main results of the research:-      the concepts of the stock market were researched and methods of its analysis were investigated;-      the evolution of short-term forecasting methods was considered;-      samples of data sets for analysis were made and vector and complex vector autoregression models were built;-      A comparative analysis of the characteristics of the models was performed and the most effective model in terms of forecasting accuracy was selected;-      the analysis of the made forecasts was carried out.The field of application of the results of the research is the economic short-term forecasting of interdependent variables to solve various problems of socio-economic profile.The scientific novelty of the study lies in the fact that the use of complex-valued autoregressions opens great opportunities for practicing analysts, reducing the computational complexity of assessing the coefficients of the model.Conclusions. The purpose of the research has been achieved - an effective model for short-term forecasting has been defined and the tasks set in the process of implementation have been solved. The scope of applicability of the results of the thesis and the scientific significance of the obtained results have been outlined. |
Databáze: | OpenAIRE |
Externí odkaz: |