Many common software questions are answered in the categories below. If you can’t find a response for yours, try checking our User Forum, or, contact HelpDesk for direct assistance

Metrix » Tuning
Q: Should you billmatch and tune at the same time?


There are cases where you might choose to establish a correlation and billmatch at the same time.

What this means is that in addition to correcting for weather (and/or user variables), Metrix will adjust your baseline based upon the deviation of the bills from the regression equation during the tuning period. The baseline equation will then become:

kWh = Coef1 * #Days + Coef2 * #CDD + offset

where the offset is the difference between the actual billed kWh in the tuning period and the prediction of kWh using the tuning equation: kWh = Coef1 * #Days + Coef2 * #CDD.

However, you should be aware that the distance from actual bills points (on the tuning graph) to the regression line, is due to factors we cannot predict with weather, number of days (and possibly any user variables you tuned with). Examples of situations that would cause points to lay far off the line would include equipment breakdowns, anomalies in facility manager's operating practices, changes in schedules (special events like plays, sports, etc.).

If you believe that these events that cause usage to lie either above or below the fit line will continue from year to year, then go ahead and tune and billmatch. You should have a good reason to believe that the deviation is reproducible. However if you do not feel comfortable making that assumption, do not tune and billmatch. Remember, if March's bill is 10% lower than the fit line during the tuning period, if you tune and billmatch, every year March's baseline will always be 10% lower than what the regression equation predicts.

I can’t find the answer to your question ? Try checking our User Forums.

Otherwise feel free to contact our Tech Support staff at (805) 329-6565, or via email at helpdesk@abraxasenergy.com.

MetrixMarketManagerReportMakerOption C