As we approach the end of the year, anticipations for the fourth-quarter US GDP are on the rise. The Capital Spectator’s updated econometric nowcast suggests a growth rate of 2.9% (real seasonally adjusted annual rate). This is a significant increase compared to the earlier estimate of 2.0% released on December 9. The government’s initial estimate for Q4 of 2013 is set to be unveiled on January 30.
While this revised nowcast shows an improvement over last month’s projections, the forecasted growth of 2.9% still lags behind the impressive 4.1% growth recorded in Q3, as reported by the Bureau of Economic Analysis last month. Nevertheless, economists indicate a positive shift in the overall economic outlook. Recent upbeat reports on jobless claims and the ISM Manufacturing Index suggest favorable trends. “The underlying trends are pointing to the economy accelerating as we move through the year,” states Joel Naroff, chief economist at Naroff Economic Advisors, in an interview with Reuters. “Conditions seem to be aligning for a very promising year ahead.”
Below is a comparison of The Capital Spectator’s current Q4 nowcast with historical data and forecasts from various sources:
Next, let’s delve into the individual nowcasts:
Here’s how the updates for Q4:2013 nowcast compare so far:
Finally, here is a brief overview of The Capital Spectator’s various nowcast methodologies:
R-4: This estimate employs a multiple regression analysis in R, utilizing historical GDP data against quarterly changes in four key economic indicators: real personal consumption expenditures (or current month retail sales until the PCE report is released), real personal income excluding government transfers, industrial production, and private non-farm payrolls. The model draws statistical relationships from the early 1970s to the present and is updated as new data becomes available.
R-10: This model also relies on a multiple regression framework based on data dating to the early 1970s and updates as new information is available. It follows a similar approach as the R-4 model but incorporates ten factors instead of four, adding the following six series:
- ISM Manufacturing PMI Composite Index
- Housing starts
- Initial jobless claims
- The stock market (S&P 500)
- Crude oil prices (spot price for West Texas Intermediate)
- The Treasury yield curve spread (10-year Note less 3-month T-bill)
ARIMA GDP: This nowcast utilizes an autoregressive integrated moving average model. This ARIMA model forecasts the change in GDP for the target quarter using historical data dating from the early 1970s. The nowcast is recalibrated as historical GDP data is revised, calculated in R using the “forecast” package.
ARIMA R-4: This model marries ARIMA estimates with regression analysis to project GDP data. It analyzes four historical datasets: real personal consumption expenditures, real personal income excluding government transfers, industrial production, and private non-farm payrolls. The historical relationships between these indicators and GDP are utilized for projections. As the actual data becomes available, it replaces ARIMA estimates, refining the nowcast.
VAR 4: This vector autoregression model seeks interdependent relationships using four data series to estimate GDP. It uses the same historical datasets found in the R-4 and ARIMA R-4 models but employs a different econometric framework. The VAR-4 nowcast is updated as new data is released, utilizing the “vars” package in R.
ARIMA R-NIPA: This model employs an autoregressive integrated moving average to predict future GDP values, drawing from four main categories in the national income and product accounts (NIPA): personal consumption expenditures, gross private domestic investment, net exports of goods and services, and government consumption expenditures and gross investment. It utilizes historical data from the early 1970s for forecasting the change in the target quarter, with updates as the historical figures are revised, also calculated in R using the “forecast” package.
