iMarketSignals – improve investment performance

We provide unbiased guidance to market direction. Our models can be classed into following  groups:

  • Leading recession indicators:
    • BCI is a Business Cycle Indicator optimised to identify a looming recession constructed from economic data series.
    • COMP is a recession indicator model constructed by combining third party leading indicators.
  • Buy and Hold investing. These models are macro by nature; trading signals from the models occur infrequently and are un-hedged – one does not have to trade constantly in and out of the markets.
    • MAC-US is a moving average crossover model for the S&P500.
    • MAC-AU is a moving average crossover model for the Australia All Ordinaries Index.
    • BVR is a bond market model based on mathematics alone.
    • Yield-Curve is a trend-following model.
    • GOLD is a rate-of-change trend-following model.
    • SILVER is a rate-of-change trend-following model.
  • Active weekly trading models based on a proprietary ranking system updated each week to generate un-biased and un-hedged buy- and sell-signals.
    • iM-BestX are weekly trading models based on the S&P 500, S&P1500 and Russell 1000 stock index series.
    • iM-Best(SPY-SH) is a market timing model, updated weekly on Sundays for the coming week.
    • iM-Best1(Sector SPDR) is a rotation system for the Select Sector SPDR® ETFs that divide the S&P500 into 9 sectors.
    • iM-Best(SSO-TLT) is a ETF switching system based on market timing.
    • iM-Best Combo3 is a combination model of Best(SPY-SH) + Best1(Sector SPDR) + Best(SSO-TLT), updated weekly on Sundays for the coming week.

The models are updated weekly and the charts depict the state of the various markets. They can be seen by registered/logged-in members.

The models have all been published. Before the models were available on the internet, weekly updates for the IBH, MAC, BVR and Yield-Curve were sent by email to subscribers for about two years – here are some of their comments:


This model and your fine work has been quite remarkable (and profitable). I have been enjoying your observations for quite some time. Job well done.
Edward Chrusciel

Your system looks pretty good. In fact, I was amazed by the high percentage winning trades and the very low drawdowns. I’ve never seen anything like that in a MA system.
Tom Swiatek

Regardless of which version one uses, it’s a powerful system. And it answers one of the most troubling worries that investors have. It efficiently gets them out of harm’s way when the market crashes.
Erik Conley

Thank you for your unique and excellent work. We will appreciate receiving your Modelling updates. Thanks!
Chuck Szkalak

The average investor reads the financial news and thinks that gives him an edge. That only tells him what everyone else knows and is worried about. We follow the excellent work of Georg Vrba, who has a top-rated coincident recession indicator as well as a successful stock/bond asset allocation model. Here is his most recent comment:  “My own composite short leading economic indicator, which has the  highest score of all indicators so far tested, does not support the notion of a recession anytime soon.” I listen to Georg, and you should too.
Jeffrey A. Miller, PhD, President NewArc Investments, Inc.

I took that S&P buy signal of yours with a little $50K futures contract and am very happy thank you.
Dwaine van Vuuren – CEO of PowerStocks

I have read and studied all your articles with interest. I have learned much from your perspectives and will continue to follow your work.
David Hamilton

Thank you for all your outstanding analysis. Let me say that I couldn’t agree more that data beats opinion.
Dave Lincoln

Georg Vrba asks whether the ECRI is still relevant. Great analysis and charts.
Jeffrey A. Miller, PhD, President NewArc Investments, Inc.


I very much liked the write up on your models, and if I may add these have a particular appeal and interest having been developed by a fellow engineer.
Paul Willis

Have seen so many manipulate data to fit their predetermined biases and love the way you methodically deconstructed the moving averages to debunk this one. It shows the difference between intellectual laziness and intellectual rigor which you fit to a tee.
Kuosen Fung, CFP®

Your work is really interesting and I’d love to receive your updates or any other information you publish. I honestly think your results indicate there is a way to time stock and bond market exposure. Most likely people will remain human and many will follow their “gut feel” or let fear or greed override any mathematical analysis. This is why your methods will most likely continue to work.
James Schwartz, CFP®

Georg Vrba, whose excellent work on recession forecasting has helped our readers, has two different market-timing methods. His most recent article explains that the next great bull market might already be here.
Jeffrey A. Miller, PhD, President NewArc Investments, Inc.

… as always love your analysis! Always eager for Friday’s just to receive your reports!
Jose R Barcelo

I agree with your viewpoint that mathematical models provide better guidance to market directions than financial experts. Thanks so much!
C.T. Wu, PhD in EECS

I just read your article on identifying recessions. Great work!
Richard G Greenwald, AAMS, CRPC

I appreciate using math to model financial behaviour and appreciate your good work.
Richard E. Hamrick, CFA®

This time it is perma-bear Albert Edwards warning investors about the “Ultimate Death Cross” taking the S&P 500 back to 666. Georg Vrba takes the analysis even further. He shows that Edwards’ prediction is almost impossible to accomplish if you actually do the math on the moving averages involved. In addition, he demonstrates that the current indicator conditions are actually bullish based on historical data.
Jeffrey A. Miller, PhD, President NewArc Investments, Inc.

I have great respect for your work and am grateful to be included on your list.
Marvin Snyder

I have your historical charts going back and it is great material. I appreciate your insights. Steve Wenstrup


I have been following your work for some time and have been very impressed.
Steve McCarthy, CPA, CFP®

I’m impressed that you provide this much analysis each week. Thank you for your insights. Your charts are beautiful & clear.
Brett Bowman

I am really impressed by your work and liked very much your last research on gold and silver. It’s good to see that you are expanding the scope of possible investments.
Nicolas Tabourdeau

Posted in homepage

iM-Trading Models:


Expect Further Losses For Stocks but 10-Year Forward Returns Look Better: Update November 2022

  • The best fit line and prediction band were re-calculated from Jan-1871 to Sep-2022. This added over 10 years of data after July-2012, the end date of the previous regression analysis.
  • The average of S&P 500 for Oct-2022 was 3,726 (20% down from Dec-2021 average of 4,675) and is 207 points higher than the corresponding re-calibrated long-term trend value of 3,519.
  • For the S&P 500 to reach the long-trend would entail only a 6% decline from the October average value, indicating that the S&P 500 is not significantly overvalued anymore.
  • The Shiller CAPE-ratio is at 27.2, only 6% higher than its 35-year moving average (MA35), currently at 25.7, forecasting a relatively high 10-year annualized real return of about 7.5%.
  • However, rising inflation with a falling CAPE-MA35 ratio, similar to what occurred in the period 1964-1973, implies very low or negative 10-year forward annualized real returns.

An Upcoming Recession is Signaled by the Forward Rate Ratio

  • Prior to recession the yield curve becomes inverted, as indicated by the Forward Rate Ratio between the 2-year and 10-year U.S. Treasury yields (FRR2-10) being less than 1.00.
  • The FRR2-10 crosses 1.000  downward signifying that US economic activity is in the boom phase of the business cycle, nearing the next recession.
  • The average lead time after FRR2-10 becomes less than 1.00 to the subsequent recession start was 14 months for the seven of the eight last recessions.

Expect Further Losses For Stocks And Very Low 10-Year Forward Returns: Update July 2022

  • The average of S&P 500 for May 2022 was 4,040 (14% down from December 2021 average) and is still 1,469 points higher than the corresponding long-term trend value of 2,571.
  • For the S&P 500 to reach the long-trend would entail a 36% decline from the May average value, possibly over a short period.
  • The Shiller CAPE-ratio is at a level of 31.0. That is 21% higher than its 35-year moving average (MA35), currently at 25.6.
  • The CAPE-MA35 ratio is at 1.21 (down from the December 2021 level of 1.51), forecasting a 10-year annualized real return of about 6.3%.
  • However, rising inflation with a falling CAPE-MA35 ratio, similar to what occurred in the period 1964-1973, implies very low or negative 10-year forward annualized real returns.
  • The historic long-term trend indicates a 10-year forward real annualized return of only 1.9% (up from the December 2021 forecast of 0.2%).

The iM-Inflation Attuned Multi-Model Market Timer

  • Investment risk can be reduced by a multi-model market timer whose many components use different and uncorrelated financial and economic data, including inflation.
  • This model seeks to determine effective asset allocation for risk-on and risk-off periods for equities considering the effect of inflation.
  • Four risk scenarios are possible: risk-on & normal-inflation, risk-on & high-inflation, risk-off & normal-inflation, and risk-off & high-inflation. Different ETF groups apply to each risk scenario.
  • From 2000 to 2022, switching accordingly between risk-related ETF groups would have produced an annualized return of about 39% versus 6.5% for buy and hold SPY.

Evaluating Popular Asset Classes for Inflation Protection

  • We tested nine asset classes which are supposed to provide protection against inflation according to an Investopedia article. The test period was from January 2005 to May 2022.
  • Investopedia provides no definition for inflationary environment, but this analysis uses the 6-month moving average of the inflation rate and the University of Michigan: Inflation Expectation© series to define it.
  • For this investigation we consider separately the inflationary periods which fall within the Risk-on and Risk-off phases for equities, as defined by the iM-Multi-Model Market Timer.
  • From the asset classes listed by Investopedia only the Vanguard Real Estate ETF (VNQ) provided some inflation protection relative to the SPDR S&P 500 ETF (SPY).
  • Better inflation protection is provided by energy sector ETFs XLE and PXE, but energy sector funds were not among the asset classes listed in the Investopedia article.