20160812Blattner Exhibit 18.pdf

Ronald L. Williams, ISB No. 3034
Williams Bradbury, P.C.
1015 W. Hays St.
Boise, ID 83 702
Telephone: [phone]
Email: [e-mail]
Attorneys for Intermountain Gas Company
BEFORE THE IDAHO PUBLIC UTILITES COMMISSION
IN THE MATTER OF THE APPLICATION OF )
INTERMOUNTAIN GAS COMPANY FOR )
THE AUTHORITY TO CHANGE ITS RATES ) Case No. INT-G-16-02
AND CHARGES FOR NATURAL GAS )
SERVICE TO NATURAL GAS CUSTOMERS )
IN THE ST A TE OF IDAHO ) ________________ )
EXHIBIT 18
• BOISE STATE UNIVERSITY
COLLEGE OF BUSINESS AND ECONOMICS
Department of Information Technology and Supply Chain Management
June 22, 2016
Ms. Lori Blattner
lntermountain Gas Company
555 South Cole Road
Boise, Idaho 83709
Dear Ms. Blattner,
At the request of Intermountain Gas Company, we have reviewed the methodology used to develop
statistical models for forecasting monthly and annual natural gas demand. measured in therms for
three different customer classes (RS-1, RS-2, and GS). In our opinion, the methods used are
appropriate and are based on sound statistical methodology. The indicator variables used in the
forecasting models for therm usage are consistent with business practice and the economic theory.
While there are alternative statistical approaches that could be employed that would be acceptable,
there is no basis for believing that alternative methods would provide superior results to those that your
method has delivered. Therefore, we are of the opinion that the methods used by lntermountain Gas
Company are an appropriate and adequate basis for weather normalization. Furthermore, the
methodology the company has used is consistent with that previously approved by the Idaho Public
Utilities Commission. Your approach follows the methodology approved by the Idaho Public Utilities
Commission in Case U-1034-134.
You tested the forecasting accuracy potential for each model. You conducted a backward forecast
to see how well the models forecast monthly and annual therm use for the years 2010-2015. You
also ran a "true forecast" test on the first four months of 2016 which were not used in developing
the model. The forecast test results demonstrate the viability of the selected models.
In summary, based on our analysis, the forecasting approach that you have used is appropriate and
the process you used to arrive at the preferred models is consistent with standard forecasting
methodology. This opinion is supported by our academic backgrounds and experiences. Patrick
Shannon holds a Ph.D. in Statistics and Quantitative Methods from the University of Oregon, has co­
authored several university textbooks including editions 1-9 of Business Statistics: A Decision­
Making Approach, and has consulted for numerous public and private sector organizations. Phillip
Fry has a Ph.D. in quantitative business analysis from Louisiana State University. He has been a co­
author on the textbook Business Statistics: A Decision-Making Approach [editions 5-9) published by
Pearson.
Sincerely,
~6-Patrick Shannon, Ph.D.
Exhibit No. 18
Case No. INT-G-16-02
L. Blattner, IGC
p. 1 of l