Faktoru modeļu agregēta un dezagregēta pieeja IKP prognožu precizitātes mērīšanā

Bessonovs, Andrejs (2010): Faktoru modeļu agregēta un dezagregēta pieeja IKP prognožu precizitātes mērīšanā. Published in: Scientific Papers University of Latvia , Vol. Vol. 7, (2010): pp. 22-33.

Abstract

The purpose of this paper is to conduct whether the disaggregated data of GDP gives us any additional information in the sense of forecasting accuracy. To test latter hypothesis author employs Stock-Watson factor model. GDP is disaggregated both on expenditure basis and on output basis. Thus both approaches should widen overlook to comparison’s capability. In order to measure forecasting accuracy root mean squared error measure was employed. Author concludes that disaggregated approach outperforms aggregated data but at very little extent. In addition, factor model showed better results in the sense of forecasting accuracy and outperformed univariate models on average by 20-30%.

• Item Type: MPRA Paper
• English Title: Measuring GDP forecasting accuracy using factor models: aggregated vs. disaggregated approach
• Language: Latvian
• Keywords: Factor model, out-of-sample forecasting, disaggregated approach, real-time database.
• Subjects: C - Mathematical and Quantitative Methods > C3 - Multiple or Simultaneous Equation Models ; Multiple Variables > C32 - Time-Series Models ; Dynamic Quantile Regressions ; Dynamic Treatment Effect Models ; Diffusion Processes ; State Space Models; 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

References

Ajevskis Viktors, Dāvidsons Gundars. Dinamisko faktoru modeļu lietojums Latvijas iekšzemes kopprodukta prognozēšanā. Latvijas Banka, 2008. Pētījums 2/2008.

Bai Jushan, Ng Serena. Determining Number of Factors in Approximate Factor Models. Econometrica, vol. 70, No. 1, 2002, p. 191–221.

Beņkovskis Konstantīns. Mēneša rādītāju izmantošana Latvijas reālā iekšzemes kopprodukta pieauguma īstermiņa prognozēšanā. Latvijas Banka, 2008. Pētījums 5/2008.

Boivin Jean, Ng Serena. Are More Data Always Better For Factor Analysis? NBER Working Paper 9829, July 2003.

Caggiano Giovanni, Kapetanios George, Labhard Vincent. Are More Data Always Better For Factor Analysis? Results For Euro Area, The Six Largest Euro Area And The UK. ECB Working Paper No. 1051, May 2009.

Dreger Christian, Schumacher Christian. Estimating large-scale factor models for economic activity in Germany: Do they outperform simpler models? HWWA discussion paper No. 199, Hamburg Institute of International Economics, 2002.

Meļihovs Aleksejs, Rusakova Svetlana. Short-Term Forecasting of Economic Development in Latvia Using Business and Consumer Survey Data. Bank of Latvia Working Paper, No. 4, 2005.

Stock James H., Watson Mark W. Diffusion Indexes. NBER Working Paper No. 6702, August 1998.

Stock James H., Watson Mark W. Macroeconomic Forecasting Using Diffusion Indexes. Journal of Business & Economic Statistics, vol. 20, No. 2, April 2002.