The November update on industrial production is scheduled for release today (9:15 am eastern) and most analysts are expecting a rebound from October's 0.4% slump. Our econometric forecasts are mixed, however, although the average of these predictions leans modestly into positive territory with a slight 0.1% increase. That's a sign for thinking that industrial production will post a gain in today's report, but probably not as much as the consensus outlook anticipates.
Here's how the numbers stack up, followed by brief definitions for the methodologies behind our projections.
VAR-11: A vector autoregression model that analyzes eleven economic series in search of interdependent relationships through history. The forecasts are run in R using the "vars" package using 30 years of historical data for the following indicators: industrial production, private non-farm payrolls, index of weekly hours worked, US stock market (S&P 500), real personal income less current transfer receipts, ISM Manufacturing Index, spot oil prices, real personal consumption expenditures, Treasury yield spread (10 year Note less 3-month T-bill), University of Michigan Consumer Sentiment Index, and housing starts.
VAR-2: A vector autoregression model that's identical to VAR-11 with the exception that the inputs are restricted to industrial production and ISM Manufacturing Index. Several studies find that the ISM data is especially valuable for anticipating industrial production and this parsimonious VAR model focuses on exploiting the historical relationship between these indicators as a forecasting tool.
ARIMA: An autoregressive integrated moving average model that analyzes industrial production index's 30-year historical record in R via the"forecast" package.
ES: An exponential smoothing model that analyzes industrial production index's 30-year historical record in R via the "forecast" package.