MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH

The present article studies a semi-parametrical method of macro-economic forecast in periods of sharp changes in economy put forward by the authors. Data blocks are chosen on the basis of clusterization methods, which are as close to the current economic conditions as possible. The key method of clu...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	V. M. Savinova, S. A. Yarushev
Formato:	article
Lenguaje:	RU
Publicado:	Plekhanov Russian University of Economics 2020
Materias:	hybrid models forecast time series arima cluster analysis method of the closest neighbor Economics as a science HB71-74
Acceso en línea:	https://doaj.org/article/561d0fae4e724ce5a71ca046eef99a63
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:561d0fae4e724ce5a71ca046eef99a63
record_format	dspace
spelling	oai:doaj.org-article:561d0fae4e724ce5a71ca046eef99a632021-11-15T05:20:49ZMODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH2413-28292587-925110.21686/2413-2829-2020-1-32-38https://doaj.org/article/561d0fae4e724ce5a71ca046eef99a632020-02-01T00:00:00Zhttps://vest.rea.ru/jour/article/view/813https://doaj.org/toc/2413-2829https://doaj.org/toc/2587-9251The present article studies a semi-parametrical method of macro-economic forecast in periods of sharp changes in economy put forward by the authors. Data blocks are chosen on the basis of clusterization methods, which are as close to the current economic conditions as possible. The key method of clusterization is the method of the closest neighbor. Time series is split into blocks and then the closest block is chosen for the last block of observation, which demonstrates the idea of coordination of directive series movements. As a basic model of forecasting the authors use ARIMA model. The authors show advantages of this approach for forecasting during the great recession – the economic slump of 2008 for such variables as inflation rate, unemployment and real private income. This method demonstrates its superiority in comparison with parametrical linear, non-linear, single-dimension and multidimension alternative methods for the period 2007–2019. The article provides calculation s obtained as a result of computer experiment using Python language for data on inflation rate and oil prices for the mentioned period. This approach in future can be used in intellectual methods of machine teaching, such as neuron networks.V. M. SavinovaS. A. YarushevPlekhanov Russian University of Economicsarticlehybrid modelsforecasttime seriesarimacluster analysismethod of the closest neighborEconomics as a scienceHB71-74RUВестник Российского экономического университета имени Г. В. Плеханова, Vol 1, Iss 1, Pp 32-38 (2020)
institution	DOAJ
collection	DOAJ
language	RU
topic	hybrid models forecast time series arima cluster analysis method of the closest neighbor Economics as a science HB71-74
spellingShingle	hybrid models forecast time series arima cluster analysis method of the closest neighbor Economics as a science HB71-74 V. M. Savinova S. A. Yarushev MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH
description	The present article studies a semi-parametrical method of macro-economic forecast in periods of sharp changes in economy put forward by the authors. Data blocks are chosen on the basis of clusterization methods, which are as close to the current economic conditions as possible. The key method of clusterization is the method of the closest neighbor. Time series is split into blocks and then the closest block is chosen for the last block of observation, which demonstrates the idea of coordination of directive series movements. As a basic model of forecasting the authors use ARIMA model. The authors show advantages of this approach for forecasting during the great recession – the economic slump of 2008 for such variables as inflation rate, unemployment and real private income. This method demonstrates its superiority in comparison with parametrical linear, non-linear, single-dimension and multidimension alternative methods for the period 2007–2019. The article provides calculation s obtained as a result of computer experiment using Python language for data on inflation rate and oil prices for the mentioned period. This approach in future can be used in intellectual methods of machine teaching, such as neuron networks.
format	article
author	V. M. Savinova S. A. Yarushev
author_facet	V. M. Savinova S. A. Yarushev
author_sort	V. M. Savinova
title	MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH
title_short	MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH
title_full	MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH
title_fullStr	MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH
title_full_unstemmed	MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH
title_sort	models of scenario forecasting of economic crises on the basis of hybrid approach
publisher	Plekhanov Russian University of Economics
publishDate	2020
url	https://doaj.org/article/561d0fae4e724ce5a71ca046eef99a63
work_keys_str_mv	AT vmsavinova modelsofscenarioforecastingofeconomiccrisesonthebasisofhybridapproach AT sayarushev modelsofscenarioforecastingofeconomiccrisesonthebasisofhybridapproach
_version_	1718428615459209216

MODELS OF SCENARIO FORECASTING OF ECONOMIC CRISES ON THE BASIS OF HYBRID APPROACH

Ejemplares similares