Previsão de variáveis macroeconômicas brasileiras usando modelos de séries temporais de alta dimensão




Previsão, Modelos de Fatores, Métodos de Shrinkage, Combinação de Previsão, Variáveis Macroeconômicas Brasileiras


Este artigo analisa o desempenho de vários modelos de fatores de alta dimensão para prever quatro variáveis macroeconômicas brasileiras, incluindo a taxa de desemprego, o índice de produção industrial, IPCA e IPC. Os fatores são extraídos de um conjunto de dados composto por 117 variáveis macroeconômicas. Técnicas de aprendizado estatístico foram aplicadas visando aumentar a performance dos modelos fatoriais. Três tipos de técnicas de aprendizado estatístico foram aplicadas: métodos de shrinkage, combinações de previsões e seleção de previsores. Os fatores são extraídos de forma supervisionada e não supervisionada. Os resultados indicam que métodos de aprendizado estatístico melhoram o desempenho preditivo das variáveis econômicas brasileiras. Além disso, a combinação de técnicas de aprendizagem estatística e supervisão fatorial produzem melhores previsões que modelos sem fatores, do que modelos fatoriais com ou sem supervisão e do que modelos que utilizam apenas o aprendizado estatístico sem supervisião dos fatores. Única exceção a estas conclusões foi a variável índice de produção industrial que foi melhor prevista pelo modelo não supervisionado de fatores.


Não há dados estatísticos.

Biografia do Autor

Rafael Barros Barbosa, Universidade Federal do Ceará

Doutor em economia pela Universidade Federal do Ceará. Especialista em Econometria, Economia da Educação e Macroeconomia.

Roberto Tatiwa Ferreira, Universidade Federal do Ceará

Doutor em Economia pela Universidade Federal do Ceará. Professor de economia lotado no departamento de economia aplicada. Professor da Pós-graduação CAEN/UFC.


Artis, Michael, Marcellino e Proietti. “Business cycles in the new EU member countries and their conformity with the euro area”. Journal of business cycle measurement and analysis 1: 7-41, 2005
Bai, J. “Inferential theory for factor models of large dimensions”. Econometrica 71:135–171, 2003.
Bai, J. and Ng, S. “Determining the number of factors in approximate factor models”. Econometrica 70: 191–221, 2002.
Bai, J. and Ng, S. “Confidence intervals for diffusion index forecasts and inference for factoraugmented regressions”. Econometrica 74 1133–1150, 2006.
Bai, J. and Ng, S. “Forecasting economic time series using targeted predictors” Journal of Econometrics 146: 304–317, 2009.
Bai, J. and Ng, S. “Boosting diffusion indices”. Journal of Applied Econometrics 4: 607–629, 2008.
Bair, Eric, Hastie, T., Paul, D. e Tibsharani, R. “Prediction by supervised principal components”. Journal of the American Statistical Association 101, no. 473, 2006.
Boivin, J., and Ng, S. “Are more data always better for factor analysis?” Journal of Econometrics 132: 169–194, 2006.
Breiman, L. “Better subset regression using the nonnegative garrote”. Technometrics 37, no. 4: p. 373- 384, 1995.
Cheng, X. and Hansen, B. “Forecasting with factor-augmented regression: A frequentist model averaging approach”. Journal of Econometrics 186: 280-293, 2015.
Dias, F., Pinheiro, M. e Rua, A. “Forecasting using targeted diffusion indexes”, Journal of Forecasting 29, no.3: 341-352, 2010.
Efron, B., Hastie, T., Johnstone, L., and Tibshirani, R. “Least angle regression”. Annals of Statistics 32: 407-499, 2004.
Eickmeier, S. and Ziegler, C. “How successful are dynamic factor models at forecasting output and inflation? a meta-analytic approach”. Journal of Forecasting 27, no.3: 237-265, 2008.
Elliot, G. e Timmermann, A. “Economic forecasting”. Princeton University Press, New Jersey, 2016.
Fernandez, C., lLey, E. e Steel, M. “Benchmark priors for Bayesian model averaging”. Journal of Econometrics 100: 381-427, 2001.
Ferreira, R., Bierens, H. e Castelar, I. “Forecasting quarterly Brazilian GDP growth rate with linear and nonlinear diffusion index models”. EconomiA 6, no.3: 261-292, 2005.
Figueiredo, F. M. R. “Forecasting Brazilian inflation using a large data set” Brazilian Central Bank, working paper series 228, 2010.
Foster, D. e George, E. “The risk inflation criterion for multiple regression” The Annals of Statistics 22: 1947-1975, 1994.
Garcia, M. Medeiros, M. e Vasconcelos, G. 2016 “Real-time inflation forecasting with high dimensional models: The case of Brazil”. XVI Encontro de Finanças, Rio de Janeiro.
Gelper, S. and Croux, C.2008 “Least angle regression for time series forecasting with many predictors”, Working paper. technical report, Katholieke Universiteit Leuven.
Hansen, B. “Least squares model averaging”. Econometrica 75: 1175–1189, 2007.
Hansen, B. “Least squares forecasting averaging”. Journal of Econometrics 146: 342–350, 2008.
Hansen, B. e Racine, J.S. “Jackknife model averaging”. Journal of Econometrics 167: 38–46, 2012.
Hansen, P. Lunde, A. e Nason, J. M. “The model confidence set”. Econometrica 79: 453-497, 2011.
Hastie, T.; Tibshirani, R. e Friedman, J. “The elements of statistical learning: data mining, inference, and prediction”. Springer-Verlag New York. 2ª ed, 2009.
Hillebrand, e.; Huang, Y.; Lee, t.; Li, C. “Using the entire yield curve in forecasting output and inflation”, Econometrics 6, no. 40, 2018.
Inoue, A., e Kilian, L. “How useful is bagging in forecasting economics time series? a case study of us cpi inflation”. J. Amer. Statist. Assoc. 103: 511–522, 2008.
Kim, H., e Swanson, N. “Forecasting financial and macroeconomic variables using data reduction methods: new empirical evidence”. Journal of Econometrics 178: 352–367, 2014.
Kim, H., e Swanson, N. “Mining big data using parsimonious factor machine learning, variable selection, and shrinkage methods”, Rutgers University, working paper, 2016.
Koop, G. and Potter, S. “Forecasting in dynamic factor models using Bayesian model averaging”. Econometrics Journal 7: 550-565, 2004.
Liu, C. E Kuo, B. “Model Averaging In Predictive Regressions”. Econometrics Journal 19, no.2: 203-231, 2016.
Kwiatkowski, D; Phillips, P.; Schmidt, P. e Shin, Y. “Testing the Null Hypothesis of Stationarity against the Alternative of a Unit Root”. Journal of Econometrics 54, no.9: 159-178, 1992.
Mallows, C.L. “Some Comments on Cp” Technometrics 15: 661–675, 1973.
Marcellino, M. A. “Comparison of Time Series Model Forecasting GDP Growth and Inflation”. Journal of Forecasting 27: 305-340, 2008.
Medeiros, M. C.; Vasconcelos, G. and Freitas, E. “Forecasting Brazilian Inflation with High-dimensional Models”. Brazilian Econometric Review 36, no 2. 2016.
Pesaran, H. and Timmermann, A. “Selection of Estimation Window in The Presence of Breaks”, Journal of Econometrics 137, no.1: 134-161, 2007.
Pesaran, H.; Petenuzzo, D. e Timmermann, A. “Forecasting Time Series Subject To Multiple Structural Breaks” Review of Economic Studies 73: 1057-1084, 2006.
Rahal, C. “Housing Market Forecasting With Factor Combinations”, Discussion Papers 15-05r, 2015, Department of Economics, University Of Birmingham.
Rossi, B. Advances in Forecasting Under Instability. In Elliott, G. and Timmermann, A., Editors, Handbook of Economic Forecasting, Volume 2b, Chapter 21: 1203-1324, 2012.
Rossi, B. e Inoue, A. “Out-of-sample Forecast Tests Robust to The Window Size Choice”. Journal of Business and Economics Statistics 30, no.3: 432-453, 2012.
Saigo, H.; Uno, T. and Tsuda, K. “Mining Complex Genotypic Features for Predicting Hiv-1 Drug Resistance”, Bioinformatics 23: 2455–2462, 2007.
Stock, J. H. and Watson, M. W. “A Comparison of Linear and Nonlinear Univariate Models for Forecasting Macroeconomic Time Series”. In: Engle, R., White, H. (Eds.), Cointegration, Causality and Forecasting: A Festschrift for Clive W.J. Granger. Oxford University Press, 1999.
Stock, J. H. and Watson, M. W. “Forecasting Using Principal Components from a Large Number of Predictors”. Journal of the American Statistical Association 97: 1167-1179, 2002.
Stock, J. H. and Watson, M. W. “Combination Forecasts of Output Growth in a Seven Country Data Set”. Journal of Forecasting 23: 405-430, 2004.
Stock, J. H. and Watson, M. W. “Implications of Dynamic Factor Models for VAR Analysis”. Nber Working Papers 11467, 2005.
Stock, J. H. and Watson, M. W. “Forecasting With Many Predictors” In Elliott, G., Granger, C., and Timmermann, A., Editors, Handbook of Economic Forecasting 1, Chapter 10: 515-554, 2006.
Stock, J. H. and Watson, M. W. “Generalized Shrinkage Methods for Forecasting Using Many Predictors” Journal of Business and Economic Statistics 30, no.4: 481-493, 2012.
Tibshirani, R. “Regression Shrinkage and Selection via the Lasso”. Journal of the Royal Statistical Society, Series B, 58 p. 267-288, 1996.
Tu, Y., and Lee, T.H. “Forecasting Using Supervised Factor Models” Journal of Management Science and Engineering 4: 12-27, 2019.
Watson, M. And Amengual, D. “Consistent Estimation Of The Number Of Dynamic Factors In A Large N And T Panel” Journal of Business and Economic Statistics 25, no. 1: 91-96, 2007.
Yin, Shou-yung, Liu, Chu-an e Lin, C. “Focused Information Criterion And Model Averaging For Large Panels With A Multifactor Error Structure”, Ideas Working Paper: 16-a016, Institute Of Economics, Academia Sinica, 2016.
Yuan, M. and Lin, Y. 2007 “On the Non-negative Garrotte Estimator.” Journal of the Royal Statistical Society 69, no.2: 143-161.
Zou, H. “The Adaptive Lasso and Its Oracle Properties” Journal of the American Statistical Association 101: 1418-1429, 2006.
Zou, H. e Hastie, T. “Regularization and Variable Selection Via The Elastic Net”, Journal of the Royal Statistical Society, Series B Vol. 67, Part 2: 301–320, 2005.
Zhang, Ke; Yin, Fan E Xiong, S. “Comparisons of Penalized Least Squares Methods by Simulations”. Working Paper, Chinese Academy Of Sciences, 2014.







Dados de financiamento