This article presents the Trader Edge aggregate neural network model forecast for the October 2016 non-farm payroll data, which is scheduled to be released tomorrow morning at 8:30 AM EDT.
Non-Farm Payroll (NFP) Model Forecast - October 2016
The Trader Edge aggregate NFP model represents the average of three neural network forecasting models, each of which employs a different neural network architecture. Unlike expert systems, neural networks use algorithms to identify and quantify complex relationships between variables based on historical data. All three models derive their forecasts from seven explanatory variables and the changes in those variables over time.
The table in Figure 1 below includes the monthly non-farm payroll data for two months: September and October 2016. The September data was released last month and the non-farm payroll data for October 2016 will be released tomorrow morning at 8:30 AM EDT.
The model forecasts are in the third data row of the table (in blue). Note that past and current forecasts reflect the latest values of the independent variables, which means that forecasts will change when revisions are made to the historical economic data.
The monthly standard error of the model is approximately 80,500 jobs. The first and last data rows of the table report the forecast plus one standard error (in green) and the forecast minus one standard error (in red), respectively. All values are rounded to the nearest thousand. If the model errors were normally distributed, roughly 16% of the observations would fall below minus one standard error and another 16% of the observations would exceed plus one standard error.
The actual non-farm payroll release for September is in the second data row of the table (in purple). The consensus estimate (reported by Briefing.com) for October 2016 is also in the second data row of the table (in purple). The reported and consensus NFP values also include the deviation from the forecast NFP (as a multiple of the standard error of the estimate). Finally, the last column of the table includes the estimated changes from September to October 2016.
The aggregate neural network model forecast for October is only 154,000, which is down materially (-32,000) from last months forecast of 186,000, reflecting a significant deterioration in the employment environment during the month of October. The Briefing.com consensus estimate for October is 175,000, which is up modestly (+19,000) from the actual September NFP data (156,000).
The actual September data was notably below the forecast (-0.37 S.E.). The consensus estimate for October is higher than the October model forecast (+0.26 S.E.). Neither September or October appear to be outliers, which reduces the likelihood of a material surprise tomorrow.
I added a new chart recently (Figure 3 below) to make it easier to observe trends in the employment environment. The blue line depicts the model forecasts (including the latest revisions) and is exactly the same as the Forecast NFP line in Figure 2 above. However, Figure 3 also contains a purple line, which shows the 12-month moving average of the NFP model forecasts.
Why plot the moving average of the model forecasts instead of the actual NFP data? Because the actual NFP data is notoriously noisy. The Forecast NFP data more accurately captures the strength of the employment environment and the stability of the data series makes it easier to observe the trend in employment.
We can use the chart below in Figure 3 in two ways to identify the trend in employment. First, we can observe the forecast NFP data relative to the moving average. Observations below the moving average indicate a weakening in employment and vice versa. Second, we can observe the slope of the moving average line. When the moving average line is downward-sloping, employment is weakening and vice versa.
As you can see from the chart in Figure 3, the slope had been negative since early 2015, but briefly stabilized and even turned positive for a few months. Unfortunately, the trend has now reversed again and is heading downward. To make matters worse, the last two NFP model forecasts were below the 12-month moving average, which is indicative of a deteriorating employment sector.
The October consensus NFP estimate is only slightly higher than the NFP model forecast, but does suggest a downward bias to tomorrow's NFP number. This would further reduce the probability of any imminent Fed intervention at their upcoming meetings.
Basic forecasting tools can help you identify unusual consensus economic estimates, which often lead to substantial surprises and market movements. Identifying such environments in advance may help you protect your portfolio from these corrections and help you determine the optimal entry and exit points for your strategies.
In the case of the NFP data, the monthly report data is highly variable and prone to substantial revisions. As a result, having an independent and unbiased indicator of the health of the U.S. job market is especially important.
Kindle Version of Brian Johnson's New In-Depth (100+ page) Options Article is Now Available on Amazon. Currently #1 Options E-Book. Print version should be available next week.
Print and Kindle Versions of Brian Johnson's 2nd Book are Available on Amazon (75% 5-Star Reviews)
Print and Kindle Versions of Brian Johnson's 1st Book are Available on Amazon (79% 5-Star Reviews)
Copyright 2016 Trading Insights, LLC. All Rights Reserved.