GDP Modelling with Factor Model: an Impact of Nested Data on Forecasting Accuracy

Bessonovs, Andrejs (2011): GDP Modelling with Factor Model: an Impact of Nested Data on Forecasting Accuracy.

Abstract

Uncertainty associated with an optimal number of macroeconomic variables to be used in factor model is challenging since there is no criteria which states what kind of data should be used, how many variables to employ and does disaggregated data improve factor model’s forecasts. The paper studies an impact of nested macroeconomic data on Latvian GDP forecasting accuracy within factor modelling framework. Nested data means disaggregated data or sub-components of aggregated variables. We employ Stock-Watson factor model in order to estimate factors and to make GDP projections two periods ahead. Root mean square error is employed as the standard tool to measure forecasting accuracy. According to this empirical study we conclude that additional information that contained in disaggregated components of macroeconomic variables could be used to enhance Latvian GDP forecasting accuracy. The efficiency gain improving forecasts is about 0.15-0.20 percentage points of year on year quarterly growth for the forecasting period 1 quarter ahead, but for 2 quarter ahead it’s about half percentage point.

• Item Type: MPRA Paper
• Language: English
• Keywords: Factor model, forecasting, nested data, RMSE.
• Subjects: C - Mathematical and Quantitative Methods > C5 - Econometric Modeling > C53 - Forecasting and Prediction Methods ; Simulation Methods; C - Mathematical and Quantitative Methods > C2 - Single Equation Models ; Single Variables > C22 - Time-Series Models ; Dynamic Quantile Regressions ; Dynamic Treatment Effect Models ; Diffusion Processes; E - Macroeconomics and Monetary Economics > E3 - Prices, Business Fluctuations, and Cycles > E37 - Forecasting and Simulation: Models and Applications

References

Bessonovs A. (2010). Agregēta un dezagregēta faktoru modeļa pieeja IKP prognožu precizitātes mērīšanā. Latvijas Universitātes raksti, 758 sējums. pp. 22-33.

Boivin, J., and Ng, S. (2006). Are more data always better for factor analysis? Journal of Econometrics, 132:169-194.

Caggiano, G., Kapetanios., G., and Labhard, V., (2009). Are more data always better for factor analysis? Results for the Euro Area, the six largest Euro Area countries and UK. ECB Working Paper, No. 1051.

Forni, M. and Reichlin, L., (1998). Let’s Get Real: a Factor-Analytic Approach to Disaggregated Business Cycle Dynamics, Review of Economic Studies 65, 453-473.

Forni, M., Hallin, M., Lippi, M. and Reichlin, L., (2001a). Coincident and Leading Indicators for the Euro Area, Economic Journal 111, C82–85.

Johnson, R., A., Wichern D., W. (2007). Applied Multivariate Statistical Analysis. Sixth edition, Pearson Prentice Hall.

Stock, J.,H., Watson, M.W., (1998). Diffusion Indexes. NBER Working Paper No. 6702.

Stock, J. H., Watson, M. W., (2002a). Macroeconomic Forecasting Using Diffusion Indexes. Journal of Business & Economic Statistics, vol. 20, No. 2.

Stock, J. H., Watson, M. W., (2002b). Forecasting Using Principal Components from a Large Number of Predictors. Journal of American Statistical Association, Vol. 97, No. 460, pp. 1167-1179.