The MTR Timing Model was updated with revised logic.  The recent market up trend did not fire a Market Up signal in the MTR-TM. This was a concern since we use this model to time our entry into the market, so we decided to take the model offline. We then started a review process and determined what needed to be revised and why the uptrend did not fire a Market Up signal. 

There were some limits coded in the model for daily market moves. This was back tested in order to handle signals in a sideways market. The logic was revised and this daily market limitation was removed. In order to reduce false signals that the classic 4% model issues, a new method was devised to determine if an x % change in the model signaled a market change in trend. This new method was coded and tested without using moving averages but percentage change in the index over x number of days. 

This change resulted in a better overall net risk adjusted returns and a higher percentage of accurate trades (now 54%) for the MTR-TM.

The system was back tested from January 1990 to December 2008. Signals fired after December 1, 2008 were "out of sample" and reflects what would have happened if this model was live on December 1, 2008. 

Recent Signals

Returns: The returns below are from 1998 - December 2008 in order to keep this information in line with the previous data posted on this site regarding MTR-TM.