MarketTrak Question/Comment Message

Posted By: Mike   Date: Thu Mar 13, 2014

Title: Intial & Final Theoretical Averages Data

  Your twitter message today provoked me to ask a few questions:
  1) What do you attribute to the big differences between the initial & final theoretical daily averages data?
  2) You mentioned the forecast for today would've been different had you waited for the final edit. Is the final edit something you would entertain doing daily instead from here on out? The extra 30 to 45 minute wait would be well worth it if it is that much more accurate & it doesn't really change a whole lot from a bid/ask spread standpoint making a trade after hours.
  Many thanks!


 Good questions Mike,

 The model has been trained using the theoretical high and low closing values and therefore expects this type of data. The theoretical values are computed after the close from the individual stocks making up the DJIA and the results are published up to several hours later. The actual high and low data are often used as an initial guess of theoretical values but because the model is so sensitive to this data, the forecast computed with actual data can be quite different from that computed with theoretical values.

 The theoretical values are usually published first by Barrons at around 3:30 PM MT. My data provider puts them in their database usually after 5:00 PM MT. Both can publish much later and sometimes not until the next day. When the theoretical results are not available, I have to use the actual data. Every now and then, the published theoretical values are in error and corrected at a later time.

 Many subscribers have asked me to publish the forecast as soon as possible after the close to allow them more time to trade in the after-hours session. For reasons above, this requirement conflicts with computing the most accurate forecast.


Return to message board


Copyright © 1996-2017 MarketTrak LLC.  All Rights Reserved.