At present, we’re happy to current a visitor contribution by Laurent Ferrara (Professor of Economics at Skema Enterprise College and Board Member of the Worldwide Institute of Forecasters).
On Friday 21 October, we organized with Catherine Doz at Paris College of Economics (PSE) a global workshop on macroeconomic nowcasting, with the assist of the PSE Chair “Measurement in Economics”. In occasions of worldwide uncertainties, it’s of essential significance to know the place we at the moment are, earlier than making an attempt to forecast the place we are going to go sooner or later. Macroeconomic nowcasting is an idea initially put ahead by Giannone et al. (2008) with the thought of optimally utilizing all the present obtainable info to evaluate financial circumstances of the present quarter, properly upfront of official figures typically launched with a lag by statistical places of work all over the world.
Many of the papers introduced throughout this workshop deal with latest strategies developed to nowcast necessary macro variables, through the use of both normal or different high-frequency variables. Various high-frequency variables are information that may be collected via numerous channels (internet sites, satellites, social networks, tensors …). They’re typically fairly giant databases, obtainable every day, not structured and with a weak signal-to-noise ratio. Subsequently, statistical/econometric strategies need to be carried out to filter out the information and extract a readable sign. The latest Covid disaster has underlined the necessity of high-frequency instruments capable of monitor in real-time financial exercise (see for instance the device put ahead by Lewis et al., 2020, on the New York Fed to intently observe U.S. financial exercise on a weekly foundation).
Modugno (Federal Reserve Board), joint with D. Cascaldi Garcia (Federal Reserve Board), T. Ferreira (Federal Reserve Board) and D. Giannone (Univ. of Washington & Amazon), introduced a brand new device developed to trace financial exercise within the euro space via Dynamic Issue Fashions estimated for every of the primary euro space nations (Germany, France and Italy). The primary thought is to account for leads and lags between enterprise cycles of the nations with a purpose to enhance nowcasting capacity. Opinion surveys are proved to be extraordinarily helpful towards this background. Weekly updates of euro space GDP can be found in real-time on the net web site euronowcast.com. The final GDP estimate for 2022q4, computed on October 28, stands at 0.2% (see Determine 1) and factors to a pointy slowdown in euro space financial exercise, according to extraordinarily excessive inflationary pressures placing progress in danger.
Determine 1: Euro space GDP progress (quarter-over-quarter) nowcast for 2022q4. Supply: euronowcast.com
Danilo Leiva-Leon (Banco de España), joint with C. Baumeister (Univ. Notre Dame) and E. Sims (Univ. Notre Dame), have constructed a brand new device capable of monitor weekly U.S. financial circumstances on the state stage. They managed to gather a public database of financial indicators on the state stage, together with labor market, mobility, actual exercise, expectations, monetary circumstances and households. Estimation of state-space fashions for every state results in a dashboard of financial circumstances for every state at numerous frequencies. All of the estimation outcomes can be found on website online fuelled by the authors. This piece of data seems extraordinarily helpful for economists concerned with monitoring U.S. financial exercise on the state stage. Be aware additionally that they compute an financial weak point index that summarizes the variety of states estimated to be in recession at every time limit (see Determine 2).
Determine 2: Financial Weak spot Index for the U.S.
Heiner Mikosch, with P. Kronenberg and S. Neuwirth (all ETH Zürich KOF), have created a device to trace financial exercise in many different European nations with different information units and numerous econometric fashions. The Nowcasting Lab of the KOF collects all of the obtainable info in real-time for all nations. Fashions are up to date on every day foundation as quickly as a brand new information level is out there. Apparently all of the nowcasting estimation stemming from all the assorted fashions can be found in a clear technique to customers, making that every person can deal with its favorite mannequin. Newest nowcasting outcomes for GDP progress charges for 2022q4 are introduced under in Determine 3.
Determine 3: Present GDP progress nowcasts for 2022q4. Supply: KOF, Nowcasting Lab
The U.S. month-to-month retail commerce index is among the most watched financial indicators by market members and economists because it displays family consumption, crucial element of U.S. GDP progress. Scott Courageous (Morning Seek the advice of) and colleagues from the Chicago Fed have developed a device to trace this index on a weekly foundation: Chicago Fed Advance Retail Commerce Abstract (CARTS). The methodology depends on a mixed-frequency Dynamic Issue Mannequin that includes each official information and high-frequency information from personal corporations that mirror credit score and debit card transactions, retail foot visitors, gasoline consumption or client sentiment. Sadly, in April 2022 the Chicago Fed has quickly paused the discharge of the CARTS index due to adjustments in personal information suppliers. The employees is at the moment engaged on a brand new model to be quickly launched.
Two different extra technical papers have been additionally introduced throughout this workshop. Gabriel Perez Quiros (Banco de España), joint with S. Delle Chiaie (ECB), query the usage of fashions that solely include high-frequency different information, as these information have a weak signal-to-noise ratio and usually do have a small pattern measurement, resulting in advanced estimation procedures. Through the use of a simulation train, they present that high-frequency different information alone result in poor nowcasting performances and an excellent technique for practitioners could be to combine each high-frequency different information (i.e. every day or weekly) and low-frequency official information (i.e. month-to-month or quarterly). Ivan Petrella (Warwick Enterprise College), joint with T. Drechsel (Univ. Maryland) and J. Antolin Diaz (London Enterprise College), put ahead a brand new mixed-frequency Dynamic Issue Mannequin for nowcasting that includes numerous necessary information options, comparable to long-run time-varying GDP progress, heterogeneous dynamics amongst variables and stochastic volatility. This latter characteristic proves extraordinarily helpful to account for big shocks in financial time collection, as these noticed in the course of the Covid-19 interval. Actual-time estimation of this mannequin factors out its capacity to effectively monitor U.S. financial exercise together with the newest interval. On this respect, the combination of high-frequency information into the mannequin proved extraordinarily helpful, along with extra normal month-to-month and quarterly macro indicators.
This publish written by Laurent Ferrara.
