混合週期資料 VAR模型을 이용한 分期 GDP成長率의 最適豫測
Forecasting GDP Growth Rates Using VAR with Mixed Frequency Data
0
조회
0
다운로드
5
인용
0
참고문헌
키워드
초록
본고에서 우리는 좀 더 정확한 분기GDP성장률 예측을 위해서 분기별 GDP성장률외에 다른 월별 경제변수도 함께 이용할 수 있는 방법을 개발하였다. 이를 위하여 우선 GDP 성장률과 기타 관련된 경제변수 모두를 포함하는 월별 VAR(vector autoregressive) 모형을 상정한 후, 이로부터 오직 분기별로만 관측되는 GDP성장률을 위한 예측식을 유도하였다. 유도된 예측식은 유한한 시차를 가지는 이동평균 오차를 갖게 되며, 따라서 차분변수를 도구변수로 하는 GMM 방법으로 쉽게 추정할 수 있다. 제시된 방법이 현실적으로 잘 적용될 수 있는가 알아보기 위하여, 실제로 한국의 분기GDP성장률 예측을 시도해 보았다. 분기별 GDP성장률 외에 월별로 관측되는 소비자물가지수(CPI) 및 여타 금융변수 등을 고려한 바, 제 금융변수들 중 월별 통화량 M1과 회사채의 신용스프레드를 각각 추가 이용할 경우 기존의 분기VAR모형이나 단일변수 AR(autoregressive) 모형 등에 비해 예측평균자승오차(forecasting MSE) 또는 중위수 오차(median bias)를 현저히 개선시킬 수 있는 것으로 나타났다.
Abstract
In this paper, we develop a new forecasting method for the quarterly GDP growth rates, which utilize other economic variables that are observed at a monthly frequency. For this, we formulate a monthly VAR model including all the relevant variables, from which a forecasting equation for the quarterly GDP growth rates is derived. The derived equation has a finite order moving average error, and therefore, can be readily estimated by GMM using lagged observations as instruments. To investigate the practical relevancy, we used the proposed method to forecast the Korean GDP growth rates. We find in particular that M1 and paper-bill credit spread, among potentially relevant financial variables that are observed monthly, improve the forecast of the Korean GDP growth rates significantly in terms of both mean squared error and median bias.
참고문헌 (26)
- 박준용, "경제시계열 분석" (2002), 경문사
- Abeysinghe, T, "Forecasting Singapore's Quarterly GDP with Monthly External Trade", 14, pp.505~513 (1998)
- Barro, "On the Determination of the Public Debt Journal of Political Economy", 87, pp.940~71 (1979)
- J.A, "Proceedings of the IEEE", 70, pp.907~939 (1982)
- Chen, B, "An Extended Yule-Walker Method For Estimating A Vector Autoregressive Model With Mixed Frequency Data", 13, pp.47~73 (1998)
- Clements, M, "Forecasting Economic Time Series" (1998), Cambridge University Press, Cambridge
- Diebold, "Journal of International Economics", 28, pp.315~32 (1990)
- Estrella, A, "Predicting U.S. Recessions: Financial Variables as Leading Indicators", 80, pp.45~61 (1998)
- Why Does the Paper-Bill Spread Predict Real Economic Activity, "edited by Stock University of Chicago Press" (1993)
- Hall, "Stochastic Implications of the Life Cycle-Permanent Income Hypothesis Journal of Political Economy", 86, pp.971~987 (1978)
- Hansen, "Large Sample Properties of Generalized Method of Moments Estimators", eco 50(metrica), pp.1029~54 (1982)
- Ingentino, R, "Using Monthly Data to Predict Quarterly Output", 3, pp.3~12 (1996)
- A, "Evidence from the Composition of External Finance", 83, pp.78~98 (1993)
- Kim, Yun-Yeong, "VAR Model Estimation Under Mixed-Frequency Data" (2007)
- R, "Forecasting with Bayesian Vector Autoregressions-Five Years of Experience Journal of Business and Economic Statistics", 4, pp.25~38 (1986)
- Lupoletti W, "Defining and Improving the Accuracy of Macroeconomic Forecasts Contributions from a VAR Model The Journal of Business", 59, pp.263~285 (1986)
- G, "The Optimal Collection of Seigniorage Journal of Monetary Economics", 20, pp.327~41 (1987)
- Miller, P, "Using Monthly Data to Improve Quarterly Forecasts", 20, pp.16~33 (1996)
- Nelson, "The Prediction Performance of the FRB-MIT-PENN Model of the U.S. Economy", 62, pp.902~917 (1972)
- Nunes, L.C, "Nowcasting Quarterly GDP Growth in a Monthly Coincident Indicator Model", 24, pp.575~592 (2005)
- "Interpreting the Evidence on Money Income Causality Journal of Econometrics", 40, pp.161~82 (1989)
- "Financial Market Variables Do Not Predict Real Activity", 36, pp.522~539 (1998)
- "Two Approaches to Macroeconomic Forecasting", 85, pp.23~40 (1999)
- "Consumption Spending and the Paper-Bill Spread: Theory Evidence", 36, pp.575~589 (1998)
- "Estimating A Multivariate ARMA Model with Mixed-frequency Data An Application to Forecasting U.S. GNP at Monthly Intervals Federal Reserve Bank of Atlanta" (1990)
- "Using Monthly Indicators to Predict Quarterly GDP", pp.1~100 (2006)