Now-casting inflation using high frequency data

Abstract

4Non-technical summary 5

Nowcasting U.S. Headline and Core Inflation

“…The Bureau of Economic Analysis (BEA) typically releases the other major measure of consumer prices, the 1 Notably, this formula is consistent with the way that the U.S. Bureau of Economic Analysis and the Blue Chip Economic Indicators survey report quarterly inflation rates. 2 Modugno (2013) also discusses the importance of high-frequency energy prices in nowcasting inflation. Stock and Watson (2003) suggest predictive content from oil prices for U.S. inflation, but these can differ from the gasoline prices we use.…”

“…In the closest related work to this paper, Modugno (2013) uses a dynamic factor model to nowcast year-over-year inflation in the headline CPI between January 2001 and December 2011, a slightly shorter sample than ours. During that sample, Modugno (2013) reports a nowcast RMSE of 0.23 percentage point on the day after the previous month's CPI is released, a 56.6 percent improvement over the 0.53 percentage point RMSEs from a random walk model in which year-over-year inflation is expected to remain unchanged from its previous reading.…”

“…The third alternative is a dynamic factor model (DFM) based on Modugno (2013) for nowcasting the U.S. CPI, and which built on earlier work by Giannone et al (2006Giannone et al ( , 2008.…”

“…Following Modugno (2013), the model combines data at the monthly, weekly, and daily 18 We tailor each set of explanatory variables Z to the inflation measure being nowcasted or forecasted and the variables available when making that nowcast or forecast. Doing so gives the MIDAS models the benefit of the deterministic model switching we propose in the baseline model.…”

A Nowcasting Model for Canada: Do U.S. Variables Matter?

“…Moreover, compared with other countries, Canada lacks some important series linked to production-namely industrial production and capacity utilization-that have been proved to be important in tracking real GDP growth for other economies given their timeliness 1 The United States (Lahiri and Monokroussos, 2013), the euro area (Angelini et al, 2010;Angelini et al, 2011;and Camacho and Perez-Quiros, 2010), France (Barhoumi et al, 2010), Ireland (D'Agostino et al, 2008and Liebermann, 2012), the Netherlands (De Winter, 2011), the Czech Republic (Arnostova et al, 2011;and Rusnák, 2013), New Zealand (Matheson, 2010), Norway (Aastveit et al, 2012;and Luciani and Ricci, 2014), Switzerland (Siliverstovs, 2012), China (Yiu andChow, 2010), Turkey (Modugno et al, 2016), Brazil (Bragoli et al, 2015), Mexico (Caruso, 2015), and Indonesia (Luciani et al, 2015). Moreover, the same framework has been applied to nowcast other variables than real GDP; see, among others D' Agostino et al (2015) for the euro area trade variables and Modugno (2013) for U.S. inflation. 2 see Fund (2015) and http : //www.tmx.com/resource/en/117…”