I introduced a simple non-farm payroll forecasting model a few months ago and revised the model last month. The revisions improved the fit and reduced the standard error. Going forward, I may also attempt to build a neural network model to forecast the monthly non-farm payroll data. This article explains the current model's forecast for the November non-farm payroll data, which will be released tomorrow morning.
I read about this indicator in an article titled "Roulette Wheels and Individual Stocks," which appeared in the December 2012 issue of Active Trader. The article was written by Richard J. Bauer, Jr. In the article, Bauer used an innovative approach to create a new type of market timing indicator based on the relative strength of an individual security versus a broad market index. The indicator shows promise as a market timing oscillator.
I read about this strategy in an article titled "The Low-Close Edge," which appeared in the December 2012 issue of Active Trader. The article was written by Nat Stewart. Stewart introduced a very simple, systematic, long-only equity strategy that generated remarkable returns, while only being invested in the market a fraction of the time. I was so surprised by the 44% return reported in 2008, that I programmed the strategy in AmiBroker and tested it myself. My results were even higher!
I introduced the topic of recession forecasting in late October with a discussion of Capital Spectator's Economic Trend index, which I used as a foundation for two new recession forecasting models. Last week I explained the potential limitations of forecasting models when faced with external shocks, including the fiscal cliff.
I have been doing some additional research on recession forecasting and have made several model enhancements and have also created a new neural network peak-trough model. I will discuss these developments in this article, but will not revisit the original research in detail. If you would like a refresher on recession diffusion indices and probit models, please use the links above to review the earlier articles.
The TRIN (also called the ARMS index) is a breadth indicator created by Richard W. Arms. It acts as an oscillator and extreme values can be used to identify overbought and oversold conditions. Unfortunately, the standard TRIN indicator has some unusual and undesirable characteristics, which makes it difficult to use in practice. This article introduces a modified version of the TRIN indicator, provides a graphical example, and includes AmiBroker code for the new indicator.
Last week I introduced two new probit models to forecast recessions and the period between the market's peak and trough associated with each recession - as defined by the National Bureau of Economic Research (NBER). In response to that article, a reader posed a timely and insightful question:
"According to the Congressional Budget Office the probability of a near term recession is directly related to the outcome of the current sequestration problem. Would the probit model address that kind of circumstance?"
While I responded to the reader's question in the comment section of the post, I felt the question was important enough to warrant a stand-alone post. Probit models are valuable tools, but all forecasting models are limited when faced with discrete, exogenous events.
Last week I wrote about using statistical tools to forecast recessions and referenced James Picerno, who provided the inspiration for this idea through his articles on the Capital Spectator Economic Trend Index (CS-ETI) and the use of probit models to estimate the probability of a recession. I used Picerno's explanatory variables as a starting point for developing two new recession forecasting models, which I will describe in this article.
U.S. Recession Risk Jumps 20% in November
I introduced the topic of recession forecasting in late October and have since developed several recession forecasting tools that I created by applying probit, logit, and neural network models to a diffusion index of economic and market-related variables. If you are new to Trader Edge and would like some additional background on these models, I encourage you to read the previous recession model articles in chronological order.
The probability that we are currently in a recession increased from 0.2% last month to 3.4% based on the November data. The probability that the S&P 500 is between its peak and trough associated with an NBER recession increased from 6.4% last month to 23.6% according to the UNIT model and from 5.1% last month to 27.4% according to the neural network model.
Continue reading →