Development of Research Methods for Statistical Dependences: Regression Models with Variable Structural Parameters
Abstract
The article describes methods for verification of a statistical model, which, firstly, is represented by time series of initial data and, secondly, is linear in the parameters being estimated. Traditionally, the use of econometric methods is based on the representation of the process under study in the form of a linear regression model. In this case, if the sets of explanatory and explained variables are represented by time series, the standard regression model allows obtaining only estimates of the structural parameters averaged over the time interval of the observed model variables. A natural generalization of the classical regression model (representing a wide class of practically important numerical problems both in the field of economic and statistical research, and in technical and other fields) is a model in which the structural parameters to be estimated on empirical data are variable in time. The article substantiates the methods of estimating the structural parameters of various types of statistical models, with reference to which the problem of estimating the dynamics of these parameters is relevant.
Keywords
econometric model,
time series,
regression with variable parameters,
regularization method,
decomposition of time series
Supporting agencies: This article was prepared using the materials developed with the financial support of the Russian Foundation for Basic Research (Project No. 16-06-00319)
About the Author
Nikolay V. Suvorov
Institute of Economic Forecasting of the Russian Academy of Sciences, Moscow Russian Federation
Russian Federation
Dr. Sci. (Econ.), Professor; Head, Laboratory of Forecasting the Dynamics and Structure of the National Economy
Russian Federation
Dr. Sci. (Econ.), Professor; Head, Laboratory of Forecasting the Dynamics and Structure of the National Economy
References
1. Suvorov N.V. Method for Constructing Regression Models with Dynamic Structural Parameters. Studies on Russian Economic Development. 2005;(4):143-155. (In Russ.)
2. Tikhonov A.N., Arsenin V.Ya. Methods for Solving Ill-Posed Problems. Мoscow: Science Publ.; 1974 (1-st edition); 1986 (2-nd edition). (In Russ.)
3. Suvorov N.V. Verification of an Econometric Model Based on a Priori Constraints on the Structural Parameters. Voprosy statistiki. 2016;(11):53-66.
4. Suvorov N.V. Current Trends and Problems of Improving Model Tools of Macroeconomic Analysis. Studies on Russian Economic Development. 2015;26(5):434-443.
5. Suvorov N.V., Balashova E.E., Davidkova O.B., Zenkova G.V. Econometric Methods for Investigating Dynamics Indicators of the Resource Intensity in the Domestic Economy (Tools and Statistical Results). Studies on Russian Economic Development. 2013;24(5):409-421.
6. Suvorov N.V. et al. Human Capital as a Factor of Russia’s Social and Economic Development. St. Petersburg: Nestor-History; 2016. (In Russ.)
7. Anchishkin A.I. Forecasting the Growth of the Socialist Economy. Moscow: Economics Publ.; 1973. (In Russ.)
8. Brown M. On the Theory and Measurement of Technological Change. Cambridge University Press; 1968. 214 p. (Russ. ed.: Braun M. Teoriya i izmerenie nauchnotekhnicheskogo progressa. Moscow: Ekonomika Publ.; 1971.)
9. Korkhin A.S. Modeling of Economic Systems with Distributed Lags. Moscow: Finance and Statistics Publ.; 1981. (In Russ.)
10. Lukashin Yu.P. Adaptive Methods of Short-Term Forecasting. Moscow: Statistics Publ.; 1981. (In Russ.)
11. Poirier D.J. The Econometrics of Structural Change: With Special Emphasis on Spline Functions. North-Holland Publ. Company; 1976. 206 p. (Russ. ed.: Puar’e D. Ekonometriya strukturnykh izmenenii. Moscow: Finansy i statistika Publ.; 1981.)
12. Speedy C.B., Brown R.F., Goodwin G.C. Control Theory. Moscow: Mir Publ.; 1974. (In Russ.)
13. Bramler K., Siffling G. Kalman-Bucy Filter. Moscow: Science Publ.; 1982. (In Russ.)
14. Durbin J., Koopman S.-J.. Time Series Analysis by State Space Methods. Oxford: Oxford University Press; 2001.
15. Koopman S.-J., Shephard N., Doornik J.A. Statistical Algorithms for Models in State Space Using SsfPack 2.2. Econometrics Journal. 1999;2(1):107-160.
16. Zivot E., Wang J. Modeling Financial Time Series with S-PLUS. New York: Springer-Verlag; 2002.
Review
For citations:
Suvorov N.V. Development of Research Methods for Statistical Dependences: Regression Models with Variable Structural Parameters. Voprosy statistiki. 2018;25(6):3-15. (In Russ.)
Views:
437
Email this article 