Divisia monetary aggregates and US GDP nowcasting

ABSTRACT

This article nowcasts US quarterly real GDP growth rate with dynamic factor model (DFM) using Divisia Monetary Aggregate Index, Divisia M1, M2, M3, and exploits information from a large, unbalanced panel data. GDP nowcasting is evaluating the current quarter GDP given the available economic data up to the point when the nowcasting is conducted. GDP data is published quarterly with a substantial lag, while many monetary and financial decisions are made at a higher frequency. Therefore, nowcasting GDP has become an increasingly important task for central banks. This article uses DFM to nowcast GDP, compares the nowcasting results from DFM with the simple sum monetary aggregate M1, M2, M3, to the Model with weighted corresponding Divisia Index, then calculates the contributions of the Divisia Monetary index to US GDP nowcasting.

KEYWORDS:

• Divisia monetary aggregates index
• nowcasting GDP
• dynamic factor model

JEL CLASSIFICATION:

• C43
• C53
• E01
• E51

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

¹ The literature has proposed frequency domain (Sargent and Sims 1977; Geweke and Singleton, 1980) and time Domain (Engle and Watson 1981; Stock and Watson 1989; Quah and Sargent 1992) methods.

² See Barnett (1978), Barnett (1980,1987) and Anderson, Jones and Nesmith (1997a, 1997b) for the detailed model description and procedure of the user cost derivation.

³ See more details at Barnett and Alkhareif (2013).Divisia Monetary Aggregates for the GCC Countries. University of Kansas Working Paper 2013.

⁴ See Barnett (1982) for a rigorous discussion on this topic, for nonmathematical explanations, see Barnett (2008).

⁵ http://www.centerforfinancialstability.org/amfm_data.php.

⁶ $r$ is the number of common factors and $n$ is the number of variables in the model.

⁷ The two tables of data description in appendix are similar to the one in Giannone Reichilin, Small 2008 paper.

⁸ https://server1.tepper.cmu.edu/barnett/divisia_data_sources.pdf.

⁹ http://research.stlouisfed.org/msi/.

¹⁰ http://www.centerforfinancialstability.org/amfm_data.php.

¹¹ http://www.philadelphiafed.org/research-and-data/real-time-centre/survey-of-professional-forecasters/.

¹² MAPFE is the average percentage of the absolute forecast error away from the actual official GDP data, it is another standard accuracy indicator of the forecasting results.

¹³ Moneyness measures the liquidity of monetary assets. Currency and coin are the most liquid (hundred percent liquid) and other monetary assets are not hundred percent liquid.

¹⁴ Prices of monetary assets are measured by the user costs or the opportunity costs of holding the asset for its liquidity services rather than investing it to obtain a much higher interest rate (Barnett 1980).