Skip to main content

Advertisement

SpringerLink
Log in

Predictive Performance of Mixed-Frequency Nowcasting and Forecasting Models (with Application to Philippine Inflation and GDP Growth)

  • Original Article
  • Published:
Journal of Quantitative Economics Aims and scope Submit manuscript

Abstract

This paper studies the comparative predictive accuracy of forecasting methods using mixed-frequency data, as applied to nowcasting Philippine inflation, real GDP growth, and other related macroeconomic variables. It focuses on variations of mixed-frequency dynamic latent factor models (DLFM for short) and Mixed Data Sampling (MIDAS) Regression. DLFM is parsimonious and dependent on a much smaller data set that needs to be updated regularly but technically and computationally more complicated, especially when there are mixed-frequency data. On the other hand, MIDAS is data-intensive but computationally more tractable. The analysis is done through a comparison of forecast performance measures (such as mean absolute prediction error) and application of statistical tests of comparative predictive accuracy and tests of forecast encompassing. Results obtained so far indicate that just about every method in the pool of forecasting methods studied performs best in some cases and worst in other cases. Thus, there is no clear winner. Furthermore, combining forecasts from the alternative methods, especially using least squares weights, improves forecast accuracy, and therefore is advocated for use in practice.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Square error loss is the standard approach in classical statistics; in many applications, more general loss functions (asymmetric, etc.) – e.g., gambling, financial investment analysis, weather forecasting, epidemiological issues (“death” much more serious consequence) may be more relevant. The methodology for comparing forecasts that we apply here can be modified to accommodate alternative loss functions.

  2. Most data are obtained from the Philippine Statistics Authority (PSA) and the Central Bank of the Philippines (BSP). Other sources include International Monetary Fund (IMF), Organization for Economic Cooperation and Development (OECD), and Federal Reserve Bank of St. Louis database FRED. For details, see Mariano & Ozmucur (2020b) Data Appendix.

  3. One possible motivation for including the three external variables as additional indicators is that the unobserved common factor has low correlation with these indicators and have significant coefficients in the estimated model. It may be possible to improve forecasting performance in a couple of directions. a. add a second unknown common factor, but initial calculations based on this show no significant improvement, b. add more explanatory (exogenous) variables that might help—like the financial conditions index, the labor market conditions index, and overseas remittances c. introduce lagged Y to capture own-time-dynamics.

  4. If the method of principal factors is used, the first step in factor analysis and principal components analysis is the same, but there is a very fundamental difference between the two methods. In factor analysis Y = AX + ε, where Y is observed, and X is unobserved. In principal components, there is not such a formal model, and X is observed, and Y is unobserved and constructed based on the proportion of total variance of the group explained by X variables. Also note that we have only one group in factor analysis and two groups in principal components analysis.

  5. See Mariano & Ozmucur (2020b) for additional references.

References

See Mariano & Ozmucur (2020b) for additional references.

  • Almon, S. 1965. The Distributed lag between capital appropriations and expenditures. Econometrica. 33 (1): 178–196.

    Article  Google Scholar 

  • Aruoba, S.B., F.X. Diebold, and C. Scotti. 2009. Real-time measurement of business conditions. Journal of Business and Economic Statistics. 27 (4): 417–427.

    Article  Google Scholar 

  • Bates, J.M., and C.W.J. Granger. 1969. The combination of forecasts. Operations Research Quarterly. 20 (4): 451–468.

    Article  Google Scholar 

  • Castle, J., D.F. Hendry, and O. Kitov. 2017. Forecasting and nowcasting macroeconomic variables: a methodological overview. In Handbook of Rapid Estimates, 2017th ed., ed. G.L. Mazzi and D. Ladiray, 53–199. European Union and the United Nations.

    Google Scholar 

  • Chong, Y.Y., and D.F. Hendry. 1986. Econometric evaluation of linear macro-economic models. Review of Economic Studies. 53 (4): 671–690.

    Article  Google Scholar 

  • Diebold, F.X., and R.S. Mariano. 1995. Comparing predictive accuracy. Journal of Business and Economic Statistics. 13 (3): 253–265.

    Google Scholar 

  • Diebold, F.X., and P. Pauly. 1987. Structural change and the combination of forecasts. Journal of Forecasting. 6 (1): 21–40.

    Article  Google Scholar 

  • Diebold, F.X., and P. Pauly. 1990. The use of prior information in forecast combination. International Journal of Forecasting. 6 (4): 503–508.

    Article  Google Scholar 

  • Doz, C., D. Giannone, and L. Reichlin. 2011. A two-step estimator for large approximate dynamic factor models based on Kalman filtering. Journal of Econometrics. 164 (1): 188–205.

    Article  Google Scholar 

  • Drachal, K. (2020). multDM: Multivariate Version of the Diebold-Mariano Test, Version 1.1.2, May 7, 2020. Retrieved May 7, 2020 from https://cran.r-project.org/web/packages/multDM/index.html.

  • Figlewski, S., and T. Urich. 1983. Optimal aggregation of money supply forecasts: accuracy, profitability and market efficiency. Journal of Finance. 38 (3): 695–710.

    Article  Google Scholar 

  • Foroni, C., and M. Marcellino. 2017. a survey of econometric methods for mixed-frequency data. In Handbook of Rapid Estimates, 2017 ed., eds. G.L. Mazzi and D. Ladiray, 251–285. European Union and the United Nations.

    Google Scholar 

  • Geweke, J. 1977. The dynamic factor analysis of economic time series. In Latent variables in socio-economic models, ed. D.J. Aigner and A.S. Goldberger, 365–383. Amsterdam: North-Holland.

    Google Scholar 

  • Ghysels, E. 2016. Macroeconomics and the reality of mixed-frequency data. Journal of Econometrics. 193 (2): 294–314.

    Article  Google Scholar 

  • Ghysels, E., and M. Marcellino. 2018. applied economic forecasting using time series methods. Oxford University Press.

    Google Scholar 

  • Ghysels, E., Santa-Clara, P., and Valkanov, R. (2002). The MIDAS touch: mixed data sampling regression models. Working paper, Chapel Hill, NC. University of North Carolina and UCLA Discussion Paper.

  • Granger, C.W.J., and R. Ramanathan. 1984. Improved methods of combining forecasts. Journal of Forecasting. 3 (2): 197–204.

    Article  Google Scholar 

  • Harvey, D., S. Leybourne, and P. Newbold. 1997. Testing the equality of prediction mean squared errors. International Journal of Forecasting. 13 (2): 281–291.

    Article  Google Scholar 

  • Hastie, T., Tibshirani, R. and Friedman, J. (2008). The Elements of Statistical Learning, Data Mining, Inference, and Prediction (2nd edition). Springer. < https://web.stanford.edu/~hastie/Papers/ESLII.pdf#page=80>

  • Hoerl, A.E., and R. Kennard. 1970. Ridge regression: biased estimation for nonorthogonal problems. Technometrics 12 (April): 55–67.

    Article  Google Scholar 

  • James, W., and Stein, C. (1961). Estimation with Quadratic Loss. Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Contributions to the Theory of Statistics (pp. 361—379). University of California Press, Berkeley, California

  • Kelly, B., and S. Pruitt. 2015. The three-pass regression filter: a new approach to forecasting using many predictors. Journal of Econometrics. 186 (2): 294–316.

    Article  Google Scholar 

  • Klein, L.R., and S. Ozmucur. 2001. The use of surveys in macroeconomic forecasting. In Macromodels 2001, ed. W. Welfe. Poland: University of Lodz.

    Google Scholar 

  • Klein, L.R., and S. Ozmucur. 2008. University of Pennsylvania high frequency macroeconomic modeling. In Econometric Forecasting and High-Frequency Data Analysis. Lecture Notes Series, Institute for Mathematical Sciences, National University of Singapore, ed. R.S. Mariano and Y.-K. Tse, 53–91. Singapore: World Scientific Publishers.

    Google Scholar 

  • Klein, L.R., and S. Ozmucur. 2009. Estimation of the us treasury yield curve at daily and intra-daily frequency. In The Making of National Economic Forecasts, ed. L.R. Klein, 265–293. Cheltenham, UK and Northampton, USA: Edward Elgar Publishing Ltd.

    Chapter  Google Scholar 

  • Klein, L.R., and J.Y. Park. 1993. Economic forecasting at high-frequency intervals. Journal of Forecasting. 12 (3–4): 301–319.

    Article  Google Scholar 

  • Klein, L.R., and J.Y. Park. 1995. The University of Pennsylvania model for high-frequency economic forecasting economic and financial modelling. Autumn 1995: 95–146.

    Google Scholar 

  • Klein, L. R., and Sojo, E. (1987). Combinations of high and low frequency data in Macroeconometric Models. Paper presented at American Economic Association Meetings in Chicago, December 1987.

  • Klein, L.R., and E. Sojo. 1989. Combinations of high and low frequency data in macroeconometric models. In Economics in Theory and Practice: An Eclectic Approach, ed. L.R. Klein and J. Marquez, 3–16. Kluwer Academic Publishers.

    Chapter  Google Scholar 

  • Makridakis, S., and M. Hibon. 2000. The M3-competition: results, conclusions and implications. International Journal of Forecasting 16 (4): 451–476.

    Article  Google Scholar 

  • Marcellino, M. 2004. Forecast pooling for short time series of macroeconomic variables. Oxford Bulletin of Economic and Statistics. 66: 91–112.

    Article  Google Scholar 

  • Mariano, R.S. 2002. Testing forecast accuracy. In A Companion to Economic Forecasting, ed. M.P. Clements and D.F. Hendry, 284–298. Blackwell. Oxford.

    Google Scholar 

  • Mariano, R.S., and Y. Murasawa. 2003. A New coincident index of business cycles based on monthly and quarterly series. Journal of Applied Econometrics. 18 (4): 427–443.

    Article  Google Scholar 

  • Mariano, R.S., and Y. Murasawa. 2010. A coincident index, common factors, and monthly real GDP. Oxford Bulletin of Economics and Statistics. 72 (1): 27–46.

    Article  Google Scholar 

  • Mariano, R.S., and S. Ozmucur. 2015. High-mixed-frequency dynamic latent factor forecasting models for GDP in the Philippines. Estudios De Economia Aplicada. 33 (2): 451–462.

    Google Scholar 

  • Mariano, R.S., and S. Ozmucur. 2018. High-mixed-frequency forecasting models for GDP and inflation. In Global Economic Modeling—A Volume in Honor of Lawrence Klein, World Scientific Publishing Co, ed. P. Pauly, 2–29. Pte. Ltd.

    Google Scholar 

  • Mariano, R.S., and S. Ozmucur. 2020b. High-mixed-frequency forecasting Methods in R -with Applications to Philippine GDP and Inflation. In Handbook of Statistics, Vol 42–Financial Macro and Micro Econometrics Using R, ed. C.R. Rao and H.D. Vinod, 185–227. Elsevier.

    Chapter  Google Scholar 

  • Mariano, R. S., and Ozmucur, S. (2020b). Predictive Performance of Mixed-Frequency Nowcasting and Forecasting Models (with Application to Philippine Inflation and GDP Growth). Penn Institute of Economic Research Working Paper. 20–029 PIER Paper Submission R8_16_20.pdf (upenn.edu).

  • Mariano, R.S., and D. Preve. 2012. Statistical tests for multiple forecast comparison. Journal of Econometrics. 169 (1): 123–130.

    Article  Google Scholar 

  • IHS Markit (2019) Eviews 11. Irvine, California. November 11, 2019.

  • Newbold, P., and C.W.J. Granger. 1974. Experience with forecasting univariate time series and the combination of forecasts. Journal of the Royal Statistical Society Series A. 137 (2): 131–146.

    Article  Google Scholar 

  • Stock, J. H., and Watson, M. W. (1989). New indexes of coincident and leading economic indicators. In O. J. Blanchard and S. Fischer (Eds), NBER Macroeconomics Annual 4 (pp. 351–409). Cambridge, Massachusetts: MIT Press. < https://www.nber.org/books/blan89-1>.

  • Stock, J.H., and M.W. Watson. 2002a. Forecasting using principal components from a large number of predictors. Journal of the American Statistical Association. 97 (460): 1167–1179.

    Article  Google Scholar 

  • Stock, J.H., and M.W. Watson. 2002b. Macroeconomic forecasting using diffusion indexes. Journal of Business and Economic Statistics. 20 (2): 147–162.

    Article  Google Scholar 

  • Stock, J.H., and M. Watson. 2004. Combination forecasts of output growth in a seven-country data set. Journal of Forecasting. 23 (6): 405–430.

    Article  Google Scholar 

  • Theil, H. 1958. Economic Forecasts and Policy. Amsterdam: North-Holland.

    Google Scholar 

  • Theil, H. 1961. Economic Forecasts and Policy, 2nd ed. Amsterdam: North-Holland.

    Google Scholar 

  • Tibshirani, R. 1996. Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society, Series B. 58 (1): 267–288.

    Google Scholar 

  • Timmermann, A. 2006. Forecast combinations. In Handbook of Economic Forecasting, vol. 1, ed. G. Elliott, C.W.J. Granger, and A. Timmermann, 135–196. Elsevier.

    Chapter  Google Scholar 

  • Zou, H., and T. Hastie. 2005. Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society Series B. 67 (2): 301–320.

    Article  Google Scholar 

Download references

Acknowledgements

The authors thank the editors and the two anonymous referees for their constructive comments and suggestions. The authors also wish to thank the University of Pennsylvania School of Arts and Sciences (SAS) Research Opportunity Grant for partial funding support for this research. An earlier and longer version of the paper was presented at the Philippine-American Academy of Science and Engineering (PAASE), July 27, 2020. Comments and suggestions of the organizers and participants of that meeting are gratefully acknowledged.

Funding

No funding.

Author information

Authors and Affiliations

  1. Department of Economics, University of Pennsylvania, 133 South 36th Street, Philadelphia, PA, 19104-6297, USA

    Roberto S. Mariano & Suleyman Ozmucur

Authors
  1. Roberto S. Mariano
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Suleyman Ozmucur
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Roberto S. Mariano.

Ethics declarations

Conflict of interest

The authors have no conflicts of interest to declare that are relevant to the content of this article.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mariano, R.S., Ozmucur, S. Predictive Performance of Mixed-Frequency Nowcasting and Forecasting Models (with Application to Philippine Inflation and GDP Growth). J. Quant. Econ. 19 (Suppl 1), 383–400 (2021). https://doi.org/10.1007/s40953-021-00276-6

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s40953-021-00276-6

Keywords

JEL Classification

Search

Navigation