The September data profile is nearly complete and the numbers provided so far reflect an economy that continues to grow. Of the 12 indicators published to date for The Capital Spectator Economic Trend Index (CS-ETI), 9 are trending positive. That’s a strong signal for assuming that recession risk was still low last month.

As the table below shows, there are relatively few signs of trouble according to a broad set of economic and financial indicators. In fact, one danger sign for July and August turned positive last month (ISM Manufacturing).

The 3-month moving average for CS-ETI, which is calculated from the numbers in the table above, is currently at 75.0% through September, based on the published data. As the next chart suggests, recession risk at these levels is a low-probability event according to the historical record. A drop below 60% would be a warning sign, and sliding under 50% would indicate that a new recession is a virtual certainty. Fortunately, CS-ETI’s 75.0% reading appears relatively stable and comfortably above the hazard zone. Although the September level to date is slightly below August’s 76.2% on a 3-month moving average basis, that’s a slight fall and well within the normal range of fluctuations during periods of economic growth.

For some perspective on CS-ETI’s outlook for the immediate future, consider ARIMA estimates to fill in the missing numbers for September, along with guesstimates for October and November. ¹ Each indicator for CS-ETI is forecast independently, with the results aggregated to estimate CS-ETI’s 3-month average. Any one forecast is likely to suffer error, of course, but predicting all the indicators in a robust econometric framework should minimize the risk a bit if some of the errors cancel each other out. Using the ARIMA estimates to fill in the gaps tells us that CS-ETI’s 3-month average for September will more or less hold steady in the 75% range once the final numbers arrive. Meantime, October and November are expected to post higher readings for CS-ETI, according to the ARIMA outlook.

1. The ARIMA forecasts are calculated in R software, using Professor Rob Hyndman’s “forecast package, which optimizes the model’s parameters based on each data set’s historical record. ^