1 OFFICIAL FILING BEFORE THE PUBLIC SERVICE COMMISSION OF WISCONSIN Application of Northern States Power Company, a Wisconsin Corporation, for Authority to Adjust Electric and Natural Gas Rates Docket No. 0-UR- DIRECT TESTIMONY OF JANNELL E. MARKS Q. Please state your full name and business address. A. My name is Jannell E. Marks. My business address is 0 Larimer Street, Suite 00, Denver, Colorado 00. Q. By whom are you employed and in what capacity? A. I am the Director of Sales, Energy and Demand Forecasting for Xcel Energy Services Inc. ( XES ), the service company for Xcel Energy Inc. ( Xcel Energy ). Q. Would you state briefly the duties of your present position? A. I am responsible for the development of forecasted sales data and economic conditions for Xcel Energy s operating companies, and the presentation of this information to Xcel Energy s senior management, other Xcel Energy departments, and externally to various regulatory and reporting agencies. I am also responsible for developing and implementing forecasting, planning, and load analysis studies for regulatory proceedings. Q. What is your educational and professional background? A. I have a Bachelor of Science degree in Statistics from Colorado State University. I began my employment with Public Service Company of Colorado ( PSCo ) in in the Economics and Forecasting Department. In, I was promoted to Senior Research Analyst and assumed responsibility for developing the customer and sales forecasts for D.
2 Docket No. 0-UR- 0 PSCo and the economic, customer, sales, and demand forecasts for Cheyenne Light, Fuel and Power Company. In, when PSCo merged with Southwestern Public Service to form New Century Energies, Inc. ( NCE ), I assumed the position of Manager, Demand, Energy and Customer Forecasts. In this position, I was responsible for developing demand, energy, and customer forecasts for NCE s operating companies. I also directed the preparation of statistical reporting for regulatory agencies and others regarding historical and forecasted reports. In August 000, following the merger of NCE and Northern States Power Company, I was named Manager, Energy Forecasting with added responsibilities for Northern States Power Company, a Wisconsin corporation and wholly owned subsidiary of Xcel Energy ( NSPW or Company ) and Northern States Power Company, a Minnesota corporation and wholly owned subsidiary of Xcel Energy ( NSPM ). I assumed my current position in February 00. Q. Have you previously testified before state regulatory agencies? A. Yes. I have previously testified before the Colorado Public Utilities Commission, the Public Utility Commission of Texas, the Minnesota Public Utilities Commission, the North Dakota Public Service Commission, the South Dakota Public Utilities Commission, the New Mexico Public Regulation Commission, and the Public Service Commission of Wisconsin ( PSCW or Commission ). Q. What is the purpose of your testimony? A. In this proceeding I am testifying on behalf of NSPW. I am sponsoring the Company s forecasts of sales and customers for the 0 test year. I recommend that the Commission adopt the Company s forecasts of sales and customers for the purpose of determining the revenue requirement and final rates in this proceeding. I present the historical customer D.0
3 Docket No. 0-UR- 0 and sales growth rates, and the megawatt-hour ( MWh) electricity sales, decatherm ( Dth ) natural gas sales, and customer forecasts for NSPW s Wisconsin retail service territory. I also present details of the methods I used to develop the MWh sales, Dth sales, and customer forecasts. Q. Are you sponsoring any exhibits in connection with your direct testimony? A. Yes. I am sponsoring Exhibit. (JEM-) consisting of seven schedules. In the remainder of my testimony all references to schedules refer to schedules of Exhibit. (JEM-). Both public and confidential versions of Schedule have been filed in this proceeding. Q. Are there defined terms you plan to use in your testimony? A. Yes. The definitions of terms that are included in my testimony are provided in Schedule. I. CUSTOMER AND SALES FORECAST Q. What geographical area do the test year sales reflect? A. My testimony and exhibits reflect energy usage and customers in NSPW s Wisconsin retail service territory. Q. How are customer and sales forecasts used in this proceeding? A. The customer and sales forecasts are used to calculate the following: ) The monthly and annual electric and gas supply requirements; ) Test year revenue under present rates; and ) Test year revenue under proposed rates. D.
4 Docket No. 0-UR- Electric Customer and Sales Forecast Q. Please describe the customer categories included in NSPW s electric customer and sales forecasts. A. The following retail customer classes comprise NSPW s electric customer and sales forecasts: Residential, Small Commercial and Industrial, Large Commercial and Industrial, Public Street and Highway Lighting, Public Authority, and Interdepartmental. Q. What is NSPW s customer forecast for the 0 test year? A. Schedule summarizes the number of electric customers for each customer class. The forecast shows, total retail customers on average for the test year. The total number of electric retail customers in NSPW s Wisconsin service territory is expected to increase by,0 customers or 0. percent over 0 levels. Q. How does the test year electric customer growth compare with historical customer growth? A. Table provides the average historical annual customer growth rates by class for the time period, as well as the forecast of annual customer growth rates by class for 0 and 0. D.
5 Docket No. 0-UR- Table Average Annual Percent Change in Electric Customers Customer Class Average Residential.0% 0.% 0.% Total Commercial & Industrial.% 0.%.0% Street Lighting.%.%.0% Public Authority -0.% -0.% -0.% Total Retail.% 0.% 0.% Q. What is NSPW s forecast of retail electric sales for the 0 test year? A. Schedule also summarizes monthly test year MWh sales for each customer class. The forecast of total retail sales in the test year is,, MWh. The forecast for annual sales growth in NSPW s Wisconsin service territory for the 0 test year is one and onehalf percent (.%). Q. How do the 0 test year electric sales compare with historical weather-normalized electric sales? A. Table provides the historical average annual electric sales growth rates by class for the period and the forecast of annual electric sales growth rates for 0 and 0. The growth rates in the 00-0 time period were heavily impacted by the economic recession, and significantly dampened the average. D.
6 Docket No. 0-UR- Table Average Annual Percent Change in Electric Sales Customer Class Average 0 0 Residential 0.% 0.% 0.% Total Commercial & Industrial.% 0.%.% Street Lighting -0.%.% 0.% Public Authority -.% -.% -.% Interdepartmental.% -.% 0.0% Total Retail 0.% 0.%.% Q. Why does the Commercial and Industrial class show a. percent increase in the 0 test year, when the average historical growth has been. percent per year? A. The. percent increase is due to a large customer, currently operating at reduced levels, but expected to return to normal operating levels in 0. If this customer s load is removed from both the 0 and 0 sales forecast, the expected growth rate in the 0 test year would be. percent for the remainder of the Commercial and Industrial class. Q. Does the sales forecast reflect expectations for an improving economy, as compared to the slow growth over the past two years resulting from the recent recession? A. Based on the economic outlook for the Eau Claire metropolitan statistical area provided by Global Insight, Inc., the economy is expected to improve over the next several years, but the rate of improvement will vary depending on the economic indicator. For example, total Historical and forecasted economic and demographic variables for Wisconsin and the Eau Claire metropolitan statistical area were obtained from Global Insight, Inc., a respected economic forecasting firm frequently relied on by forecasting professionals. These variables include population, households, employment, personal income, and Gross Metropolitan Product. The Company considers Eau Claire to be representative of the entire service territory. D.
7 Docket No. 0-UR- 0 employment levels in the Eau Claire metropolitan area peaked in the second quarter of 00, at just below,000 employees. By the end of 00, employment levels had declined by nearly four percent and 00 saw an additional three percent loss in jobs. Employment levels have been increasing since the second quarter of 0. However, even with two percent growth expected for 0, and greater than two percent growth expected for both 0 and 0, employment is not expected to reach pre-recessionary levels until the fourth quarter of 0. Another economic indicator, real gross metropolitan product for the Eau Claire region, peaked in the second quarter of 00, and then declined on an annualized basis until the first quarter of 0. By the end of 0, real gross metropolitan product had returned to the pre-recessionary levels, and year-over-year growth is expected to continue at a moderate rate in 0 and 0, at less than three percent per year. Overall, the economic outlook calls for a slow but steady recovery over the next few years. The customer and sales forecasts in this proceeding reflect this outlook, as they are based on these economic indicators. Gas Customer and Sales Forecast Q. Please describe the customer categories included in NSPW s gas customer and sales forecasts. A. The following retail customer classes comprise NSPW s gas customer and sales forecasts: Residential, Firm Commercial (or Small Commercial or General Service), Large General Service (or Demand), Firm Interdepartmental Sales, Generation Sales, Small Volume Interruptible, Medium Volume Interruptible, Firm Transportation, Interruptible Transportation, and Generation Transportation (or Interdepartmental Transportation). D.
8 Docket No. 0-UR- Q. What is NSPW s natural gas customer forecast for the 0 test year? A. Schedule summarizes the number of gas customers for each customer class. The forecast shows, total gas customers on average for the test year. The total number of gas customers in NSPW s Wisconsin service territory is expected to increase by, customers or. percent over 0 levels. Q. How does the test year gas customer growth compare with historical customer growth? A. Table provides the historical average annual retail gas customer growth rates by class for and the forecast of annual retail gas customer growth rates by class for 0 and 0. Table Average Annual Percent Change in Retail Gas Customers Customer Class Average Residential.%.%.% Total Firm Commercial.%.0%.% Total Interruptible -.% -.% -.% Total Retail.%.%.% Q. What is NSPW s forecast of natural gas sales in the 0 test year? A. Schedule also summarizes the monthly test year Dth sales for each customer class. The forecast of total sales in the test year is,, Dth. Q. How do the 0 test year gas sales compare with historical weather-normalized gas sales? D.
9 Docket No. 0-UR- A. Table provides the average annual percent change in gas sales and transportation volumes for the past ten years and the forecast for 0 and 0. Table Average Annual Percent Change in Gas Sales Customer Class Average 0 0 Residential 0.% 0.%.0% Total Commercial 0.% 0.%.% Total Interruptible -.% 0.% -.% Total Retail excl Interdepartmental -.% 0.%.% Total Interdepartmental and Transportation.% -.% -0.% Total Volumes -0.% -.% 0.% Q. Why does the Commercial class show a. percent increase in the 0 test year, when the average historical growth has been 0. percent per year? A. The. percent increase is due to the addition of a new large customer during the fourth quarter of 0. If this customer s additional load is removed from both the 0 and 0 sales forecast, the expected growth rate in the 0 test year would be 0. percent for the remainder of the Commercial class. II. OVERVIEW OF SALES AND CUSTOMER FORECASTING METHODOLOGY Q. Please describe in general terms the methods used to forecast sales and customers. A. The electric and natural gas sales and customer-forecasts are prepared using a combination of econometric and statistical forecasting techniques and analyses, including regression models and trend analysis. D.
10 Docket No. 0-UR- 0 Electric Sales and Customer Forecasting Methodology Q. How were the electric retail sales forecasts developed for the Residential, Small Commercial and Industrial, Large Commercial and Industrial, and Public Street and Highway Lighting customer classes? A. Ordinary Least Squares ( OLS ) multiple regression models were used as the foundation of the electric sales forecasts for the Residential, Small Commercial and Industrial, Large Commercial and Industrial, and Public Street and Highway Lighting customer classes. OLS multiple regression techniques are very well-known, proven methods of forecasting and are commonly accepted by forecasters and regulators throughout the utility industry. This method provides reliable, accurate projections, accommodates the use of predictor variables, such as economic or demographic indicators and weather, and allows clear interpretation of the model. Monthly sales forecasts for these customer classes were developed based on OLS regression models designed to define a statistical relationship between the historical sales and the independent predictor variables, including historical economic and demographic indicators, historical weather (expressed in heating degree days and temperature-humidity index ( THI )), and historical number of customers. Once the historical relationship was defined, the forecast was simulated using normal weather (expressed in terms of 0-yearaveraged heating-degree days and THI) and projected levels of the economic and demographic indicators. Q. What process was used to forecast electric sales in the Public Authority and Interdepartmental customer classes? D.
11 Docket No. 0-UR- 0 A. Sales in these two retail classes represent a very small proportion of total NSPW retail sales: 0. percent of total retail electric sales for the Public Authority class and only 0.0 percent for the Interdepartmental class. Usage in these classes is impacted by factors that are difficult to capture in an OLS multiple regression model, so other types of statistical analyses were used to develop the sales forecasts. For the Public Authority class, the monthly sales forecast was developed based on an extrapolation of historical growth trends. The forecast of monthly Interdepartmental sales was calculated as the average of the actual sales for each month over the past ten years. Q. What process was used for forecasting the number of electric customers? A. The number of electric customers by customer class for the classes Residential, Small Commercial and Industrial, and Public Street and Highway Lighting was forecasted using demographic data in OLS regression models. The customer forecast for the Large Commercial and Industrial customer class was developed by calculating the average number of customers in this class in 0 and holding that number constant throughout the forecast period. The forecast of the number of customers in the Public Authority class was developed by analyzing historical growth trends. Gas Sales and Customer Forecasting Methodology Q. How were the gas sales forecasts developed for the Residential, Small Commercial, and Small Interruptible customer classes? A. Monthly sales forecasts for these customer classes were developed using OLS regression models based on historical weather (expressed in heating degree days), and historical number of customers. In addition, the Residential and Small Commercial model included a predictor variable to capture the declining trend in use per customer, and the Residential D.
12 Docket No. 0-UR- 0 model included a gas price variable. Once the historical relationship was defined, the forecast was simulated using normal weather (expressed in terms of 0-year-averaged heating-degree days) and the projected levels of the other predictor variables. Q. What process was used to forecast gas sales in the Demand, Medium Volume Interruptible, Interdepartmental and Transportation customer classes? A. Usage in these classes is impacted by factors that are difficult to capture in an OLS multiple regression model, so other types of statistical analyses were used to develop the sales forecasts. For the Demand and Interdepartmental sales classes and the Firm and Interruptible Transportation classes, the monthly sales forecasts were developed based on average monthly sales over the past two years. The monthly Medium Volume Interruptible class forecast was based on average monthly sales over the past three years. Q. How was the forecast of gas for electric generation volumes developed? A. The gas for Generation forecast is an output from the Company s electric production cost model developed by Mr. Horneck. Q. What process was used for forecasting the number of gas customers? A. The forecast of the number of gas customers for the Residential and Small Commercial classes were developed using demographic data in OLS regression models. The customer forecast for the Small Volume Interruptible customer class was developed by analyzing historical customer count trends. The forecasts of gas customers for the Demand, Medium Volume Interruptible, and Transportation were developed by holding the number constant at the December 0 level. D.0
13 Docket No. 0-UR- 0 III. STATISTICALLY MODELED FORECASTS Q. Please describe the regression models and associated analysis used in NSPW s statistical projections of sales and customers. A. The regression models and associated analysis used in NSPW s statistical projections of electric sales are provided in Schedule, and the regression models and associated analysis used in NSPW s statistical projections of electric customers are provided in Schedule. The regression models and associated analysis used in NSPW s statistical projections of gas sales are provided in Schedule, and the regression models and associated analysis used in NSPW s statistical projections of gas customers are provided in Schedule. These schedules include, by customer class, the models with their summary statistics and output and descriptions for each variable included in the model. Q. What techniques did NSPW employ to evaluate the validity of its quantitative forecasting models and sales projections? A. There are a number of quantitative and qualitative validity tests that are applicable to OLS multiple regression analyses. The coefficient of determination ( R-squared ) test statistic is a measure of the quality of the model s fit to the historical data. It represents the proportion of the variation of the historical sales around their mean value that can be attributed to the functional relationship between the historical sales and the explanatory variables included in the model. If the R-squared statistic is high, the model is explaining a high degree of the historical-sales variability. The regression models used to develop the sales and customer forecasts for the 0 test year demonstrate very high R-squared statistics ranging between 0. and 0.. D.
14 Docket No. 0-UR- 0 The t-statistics of the explanatory variables indicate the degree of correlation between that variable s data series and the sales data series being modeled. The t-statistic is a measure of the statistical significance of each variable s individual contribution to the prediction model. Generally, the absolute value of each t-statistic should be greater than.0 to be considered statistically significant at the percent confidence level, and greater than. to be considered statistically significant at the 0 percent confidence level. This criteria was applied in the development of the regression models used to develop the sales forecast. The final regression models used to develop the sales and customer forecasts tested satisfactorily under this standard, with the majority of the explanatory variables being significant at the percent confidence level. Each model was inspected for the presence of first-order autocorrelation, as measured by the Durbin-Watson ( DW ) test statistic. Autocorrelation refers to the correlation of the model s error terms for different time periods. For example, an overestimate in one period is likely to lead to an overestimate in the succeeding period, and vice versa, under the presence of first-order autocorrelation. Thus, when forecasting with an OLS regression model, absence of autocorrelation between the residual errors is very important. The DW test statistic ranges between 0 and and provides a measure to test for autocorrelation. In the absence of first-order autocorrelation, the DW test statistic equals.0. The final regression models used to develop the sales forecast tested satisfactorily for the absence of first-order autocorrelation with the DW test statistics ranging between. and., which is within the percent significance level. Graphical inspection of each model s error terms (i.e. actual less predicted) was used to verify that the models were not misspecified, and that statistical assumptions pertaining D.
15 Docket No. 0-UR- 0 to constant variance among the residual terms and their random distribution with respect to the predictor variables were not violated. Analysis of each model s residuals indicated that the residuals were homoscedastic (constant variance) and randomly distributed, indicating that the OLS linear regression modeling technique was an appropriate selection for each customer class sales that were statistically modeled. The statistically modeled sales and customer count forecasts for each customer class have been reviewed for reasonableness as compared to the respective monthly sales and customer count history for that class. Graphical inspection reveals that the patterns of the forecast fit well with the respective historical patterns for each customer class. The annual total forecast sales and customer counts have been compared to their respective historical trends for consistency. Q. How accurate have the Company s forecasts of Wisconsin retail sales and customer counts been historically? A. The historical forecasts of electric sales have been within +/-. percent of actual levels over the last five years after adjusting for weather, with the average variance being less than -0. percent. The historical forecasts of number of electric customers have been within +/-. percent of actual levels, with the average variance being less than -0. percent. The historical forecasts of gas sales have been within +/-. percent of actual levels over the last five years after adjusting for weather, with the average variance being less than -0. percent. The historical forecasts of number of gas customers have been within +/-. percent of actual levels, with the average variance being less than -. percent. Given these statistics the Company has an extremely high level of confidence in the 0 test year forecasts. D.
16 Docket No. 0-UR- 0 IV. TEST YEAR SALES FORECAST ADJUSTMENTS Q. How did NSPW adjust its test year sales forecast for the influence of weather on sales? A. The classes that exhibit weather sensitivity are electric Residential, electric Small Commercial and Industrial, gas Residential, gas Small Commercial, and gas Small Interruptible. The sales projections for these classes were developed through the application of quantitative statistical models. For each of these classes, sales were not weather-adjusted prior to developing the respective statistical models. The respective linear-regression models used to forecast sales included monthly weather, as measured in terms of heating-degree days and/or temperature-humidity index ( THI ), as explanatory variables. In this way, the historical-weather impact on historical consumption for each class was modeled through the respective coefficients for the heating-degree day and THI variables included in each class model. Test year sales were then projected by simulating the established statistical relationships over the forecast horizon and assuming normal weather. For all other classes, forecast volumes have not been weather normalized. These customers use of electricity is influenced by factors other than weather, and, as a result, the weather impact due to deviation from normal weather is indistinguishable from other variables. Q. How was normal weather determined? A. Normal daily weather was calculated based on the average of historical heating-degree days and THI for the 0-year time period to 0. These normal heating-degree days D.
17 Docket No. 0-UR- 0 and THI were related to the forecasted billing month in the same manner as were the actual heating-degree days and THI, using meter reading schedules. Q. What other adjustments were made to the test year sales forecast? A. Adjustments were made to the test year sales forecast to reflect unbilled sales (i.e. sales consumed in one month that are not billed to the customer until the succeeding month), and to convert projected billing-month sales to calendar-month sales. The purpose of these adjustments is to align the test year sales and revenues with the relevant projected test year expenses. V. CONCLUSION Q. Please summarize your testimony? A. I have presented the Company s forecasts of electric and gas sales and customers for the 0 test year. I have described the historical customer and sales trends and presented details of the methods I used to develop the sales and customer forecast and the results. Q. In your opinion, does the NSPW sales and customer forecast provide a reasonable basis for establishing rates in the case? A. Yes. The forecast data is a reasonable estimate of test year sales volumes and customer counts. The forecasts are derived from robust models that are demonstrated to be highly accurate historically. I recommend that the Commission adopt NSPW s forecasts of sales and customers, as reflected in Schedules and for the purpose of determining the revenue requirement and final rates in this proceeding. Q. Does this conclude your testimony? A. Yes, it does. D.
I. FORECAST METHODOLOGY This Appendix provides a discussion of the methodology used in Resource Planning to forecast customer need, first at a summary level, and in Section II, we provide additional technical
Case No. 0-E- NEW YORK STATE ELECTRIC & GAS CORPORATION DIRECT TESTIMONY OF THE SALES AND REVENUE PANEL September 0, 00 Patricia J. Clune Michael J. Purtell 0 Q. Please state the names of the members on
ARKANSAS PUBLIC SERVICE COMMISSYF cc7 L I :b; -Ir '3, :I: 36 DOCKET NO. 00-1 90-U 1.. T -3. - " ~..-.ij IN THE MATTER OF A PROGRESS REPORT TO THE GENERAL ASSEMBLY ON THE DEVELOPMENT OF COMPETITION IN ELECTRIC
Methodology For Illinois Electric Customers and Sales Forecasts: 2016-2025 In December 2014, an electric rate case was finalized in MEC s Illinois service territory. As a result of the implementation of
Connexus Energy Ramsey, MN Measurement and Verification Report of OPower Energy Efficiency Pilot Program July 28, 2010 Contact: Chris Ivanov 1532 W. Broadway Madison, WI 53713 Direct: 608-268-3516 Fax:
BEFORE THE LONG ISLAND POWER AUTHORITY ------------------------------------------------------------ IN THE MATTER of a Three-Year Rate Plan Case -00 ------------------------------------------------------------
Integrated Resource Plan March 19, 2004 PREPARED FOR KAUA I ISLAND UTILITY COOPERATIVE LCG Consulting 4962 El Camino Real, Suite 112 Los Altos, CA 94022 650-962-9670 1 IRP 1 ELECTRIC LOAD FORECASTING 1.1
BEFORE THE PUBLIC UTILITIES COMMISSION OF THE STATE OF COLORADO * * * * * IN THE MATTER OF THE APPLICATION OF PUBLIC SERVICE COMPANY OF COLORADO FOR APPROVAL OF ITS 01 RENEWABLE ENERGY STANDARD COMPLIANCE
Not Your Dad s Magic Eight Ball Prepared for the NCSL Fiscal Analysts Seminar, October 21, 2014 Jim Landers, Office of Fiscal and Management Analysis, Indiana Legislative Services Agency Actual Forecast
CHAPTER 2 LOAD 2.1 METHODS The Company uses two econometric models with an end-use orientation to forecast energy sales. The first is a customer class level model ( sales model ) and the second is an hourly
DOCKET NO. 00 APPLICATION OF SOUTHWESTERN PUBLIC SERVICE COMPANY FOR AUTHORITY TO CHANGE RATES AND TO RECONCILE FUEL AND PURCHASED POWER COSTS FOR THE PERIOD JULY 1, 01 THROUGH JUNE 0, 01 PUBLIC UTILITY
Econometric Modelling for Revenue Projections Annex E 1. An econometric modelling exercise has been undertaken to calibrate the quantitative relationship between the five major items of government revenue
A Primer on Forecasting Business Performance There are two common approaches to forecasting: qualitative and quantitative. Qualitative forecasting methods are important when historical data is not available.
PRODUCTION PLANNING AND CONTROL CHAPTER 2: FORECASTING Forecasting the first step in planning. Estimating the future demand for products and services and the necessary resources to produce these outputs
BEFORE THE NEW MEXICO PUBLIC REGULATION COMMISSION IN THE MATTER OF SOUTHWESTERN PUBLIC SERVICE COMPANY S APPLICATION FOR APPROVAL OF ITS 00 ENERGY EFFICIENCY AND LOAD MANAGEMENT PLAN AND ASSOCIATED PROGRAMS
BEFORE THE PUBLIC UTILITIES COMMISSION OF THE STATE OF COLORADO * * * * IN THE MATTER OF THE APPLICATION OF PUBLIC SERVICE COMPANY OF COLORADO FOR APPROVAL OF ITS 0 ELECTRIC RESOURCE PLAN ) ) ) ) DOCKET
Q. PLEASE STATE YOUR NAME AND BUSINESS ADDRESS. A. My name is Byron L. Harris. My present business address is Kanawha Boulevard, East, 00 Union Building, Charleston, West Virginia 01. Q. PLEASE STATE YOUR
OFFICIAL FILING BEFORE THE PUBLIC SERVICE COMMISSION OF WISCONSIN Application of Northern States Power Company, a Wisconsin Corporation, for Authority to Adjust Electric and Natural Gas Rates DIRECT TESTIMONY
Pty Ltd ABN 85 082 464 622 Level 2 / 21 Kirksway Place Hobart TAS 7000 www.auroraenergy.com.au Enquiries regarding this document should be addressed to: Network Regulatory Manager Pty Ltd GPO Box 191 Hobart
16 : Demand Forecasting 1 Session Outline Demand Forecasting Subjective methods can be used only when past data is not available. When past data is available, it is advisable that firms should use statistical
State Electricity Price Forecasting Model Summary Contents SUFG State Electricity Price Forecasting Models... 2 United States Census Regional Division Map... 3 Forecasting Results Comparison (Annual Compound
Itron, Inc. 12348 High Bluff Drive, Suite 210 San Diego, CA 92130-2650 858-724-2620 September 16, 2014 For the third year, Itron surveyed energy forecasters across North America with the goal of obtaining
April 28, 2014 REPORT #E14-281 2013 Energy Results for the Commercial Real Estate Cohorts Prepared by: Cadmus 720 SW Washington St. Portland, OR 97205 Northwest Energy Efficiency Alliance PHONE 503-688-5400
Winter Impacts of Energy Efficiency In New England April 2015 Investments in electric efficiency since 2000 reduced electric demand in New England by over 2 gigawatts. 1 These savings provide significant
2013 Economic and Revenue Forecast Update for the City of Omaha, Nebraska Kenneth A. Kriz Regents Distinguished Professor of Public Finance Director, Kansas Public Finance Center July 20, 2013 Background
Regional Employment and Unemployment Comparison Spring 2008 Eric Jamelske, Ph.D., Associate Professor Department of Economics, University of Wisconsin Eau Claire Assisted By Casey Kettler and Matthew Pehler
ELECTRICITY DEMAND IN DARWIN ( 1990-1994 ) A dissertation submitted to the Graduate School of Business Northern Territory University by THANHTANG In partial fulfilment of the requirements for the Graduate
APPENDIX H Demand Forecasts Contents 1. Overview... H-1 2. Methodology... H-2 3. Key Assumptions... H-12 4. Electric and Gas Demand Forecasts... H-16 Demand forecasts are an estimate of how much energy
Rebuttal Testimony and Schedules George E. Tyson Before the Minnesota Public Utilities Commission State of Minnesota In the Matter of the Application of Northern States Power Company for Authority to Increase
BEFORE THE STATE OF NEW YORK PUBLIC SERVICE COMMISSION In the Matter of Consolidated Edison Company of New York, Inc. Cases -E-000 and -G-00 May 0 Prepared Testimony of: Staff Management Audit Panel Henry
Multiple Linear Regression in Data Mining Contents 2.1. A Review of Multiple Linear Regression 2.2. Illustration of the Regression Process 2.3. Subset Selection in Linear Regression 1 2 Chap. 2 Multiple
Penelec Statement No. 5 BEFORE THE PENNSYLVANIA PUBLIC UTILITY COMMISSION PENNSYLVANIA ELECTRIC COMPANY DOCKET NO. R-2016-25752 Direct Testimony of Jeffrey L. Adams List of Topics Addressed Cash Working
Simple Methods and Procedures Used in Forecasting The project prepared by : Sven Gingelmaier Michael Richter Under direction of the Maria Jadamus-Hacura What Is Forecasting? Prediction of future events
Index Section A. Planning, Budgeting and Forecasting Section A.2 Forecasting techniques... 1 EduPristine CMA - Part I Page 1 of 11 Section A. Planning, Budgeting and Forecasting Section A.2 Forecasting
CENTER FOR BUSINESS & ECONOMIC RESEARCH LEE BUSI N ESS SCHOOL SOUTHERN NEVADA QUARTERLY ECONOMIC FORECAST Quarter 2 2014 Southern Nevada Economy to Continue Growth in 2014 and 2015 S ince the end of the
Short-Term Energy Outlook Supplement: Summer 2013 Outlook for Residential Electric Bills June 2013 Independent Statistics & Analysis www.eia.gov U.S. Department of Energy Washington, DC 20585 This report
DOCKET NO. 42004 APPLICATION OF SOUTHWESTERN PUBLIC SERVICE COMPANY FOR AUTHORITY TO CHANGE RATES AND TO RECONCILE FUEL AND PURCHASED POWER COSTS FOR THE PERIOD JANUARY 1, 2010 THROUGH JUNE 30, 2012 PUBLIC
3. Load Analysis and Forecasting Ameren Missouri 3. Load Analysis and Forecasting 1 Highlights Ameren Missouri expects energy consumption to grow 23% and peak demand to grow 18% over the next 20 years.
Resume of Michael J. McFadden MICHAEL J. MCFADDEN AREAS OF QUALIFICATION Rates, regulatory affairs, strategic planning, electric and gas utility operations, corporate finance, financial analysis, asset
Do Electricity Prices Reflect Economic Fundamentals?: Evidence from the California ISO Kevin F. Forbes and Ernest M. Zampelli Department of Business and Economics The Center for the Study of Energy and
Is the Forward Exchange Rate a Useful Indicator of the Future Exchange Rate? Emily Polito, Trinity College In the past two decades, there have been many empirical studies both in support of and opposing
Business Course Text Bowerman, Bruce L., Richard T. O'Connell, J. B. Orris, and Dawn C. Porter. Essentials of Business, 2nd edition, McGraw-Hill/Irwin, 2008, ISBN: 978-0-07-331988-9. Required Computing
THIS PAPER IS NOT TO BE REMOVED FROM THE EXAMINATION HALLS University of London BSc Examination 2012 BA1040 (BBA0040) +Enc Business Administration Business Statistics Date tba: Time tba DO NOT TURN OVER
Transmission Leadership Teresa Mogensen Vice President, Transmission 1 Safe Harbor This material contains forward-looking statements that are subject to certain risks, uncertainties ies and assumptions.
MULTIPLE REGRESSION AND ISSUES IN REGRESSION ANALYSIS MSR = Mean Regression Sum of Squares MSE = Mean Squared Error RSS = Regression Sum of Squares SSE = Sum of Squared Errors/Residuals α = Level of Significance
Electricity Prices Panel Presentation to CCRE Roundtable Hockley Valley March 27, 214 Amir Shalaby, Vice President, Power System Planning March 27, 214 Today s Journey to Residential Bills Part I: Electricity
STATE REVENUE FORECASTING PROJECT 2010 TECHNICAL MEMORANDUM December 2, 2010 By Paul Shinn, Consultant Introduction This memorandum provides technical details and historical background for Oklahoma Policy
STATE OF INDIANA INDIANA UTILITY REGULATORY COMMISSION PETITION OF NORTHERN INDIANA PUBLIC SERVICE COMPANY FOR APPROVAL OF ELECTRIC ENERGY EFFICIENCY PROGRAMS TO BE EFFECTIVE FOR THE PERIOD JANUARY, 06
Company: San Diego Gas & Electric Company (U 902 M) Proceeding: 2016 General Rate Case Application: A.14-11- Exhibit: SDG&E-31 SDG&E DIRECT TESTIMONY OF KENNETH E. SCHIERMEYER (ELECTRIC CUSTOMERS AND SALES)
BEFORE THE PUBLIC UTILITIES COMMISSION OF THE STATE OF COLORADO * * * * * RE: THE TARIFF SHEETS FILED BY PUBLIC SERVICE COMPANY OF COLORADO WITH ADVICE NO. 0 GAS RE: THE TARIFF SHEETS FILED BY PUBLIC SERVICE
Elements of Revenue Forecasting II: the Elasticity Approach and Projections of Revenue Components Fiscal Analysis and Forecasting Workshop Bangkok, Thailand June 16 27, 2014 Joshua Greene Consultant IMF-TAOLAM
Integrating Financial Statement Modeling and Sales Forecasting John T. Cuddington, Colorado School of Mines Irina Khindanova, University of Denver ABSTRACT This paper shows how to integrate financial statement
1 Example G Cost of construction of nuclear power plants Description of data Table G.1 gives data, reproduced by permission of the Rand Corporation, from a report (Mooz, 1978) on 32 light water reactor
Integrating Energy Efficiency into Utility Load Forecasts Shawn Enterline, Vermont Energy Investment Corporation Eric Fox, Itron Inc. ABSTRACT Efficiency Vermont s efficiency programs are being integrated
Analytical Test Method Validation Report Template 1. Purpose The purpose of this Validation Summary Report is to summarize the finding of the validation of test method Determination of, following Validation
Simple linear regression Introduction Simple linear regression is a statistical method for obtaining a formula to predict values of one variable from another where there is a causal relationship between
APSC FILED Time: 2/19/2015 10:00:04 AM: Recvd 2/19/2015 9:51:40 AM: Docket 15-012-tf-Doc. 1 February 19, 2015 Mr. Dallas Heltz Secretary of the Commission Arkansas Public Service Commission 1000 Center
UNDERSTANDING AND MOTIVATING ENERGY CONSERVATION VIA SOCIAL NORMS Project Report: 2004 FINAL REPORT Robert Cialdini, Ph.D. Arizona State University Wesley Schultz, Ph.D. California State University, San
The Net impact of Home Energy Feedback Devices Brien Sipe, Energy Trust of Oregon, Portland OR Sarah Castor, Energy Trust of Oregon, Portland OR ABSTRACT Recently, much attention has been focused on modifying
ANNUITY LAPSE RATE MODELING: TOBIT OR NOT TOBIT? SAMUEL H. COX AND YIJIA LIN ABSTRACT. We devise an approach, using tobit models for modeling annuity lapse rates. The approach is based on data provided
VI. Real Business Cycles Models Introduction Business cycle research studies the causes and consequences of the recurrent expansions and contractions in aggregate economic activity that occur in most industrialized
VELCO LONG-TERM ENERGY AND DEMAND FORECAST FORECAST DRIVERS AND ASSUMPTIONS May 22, 2014 Eric Fox and Oleg Moskatov AGENDA» Review customer and system usage trends in Vermont» Review and discuss the forecast
BEFORE THE PUBLIC UTILITIES COMMISSION OF THE STATE OF COLORADO * * * * * IN THE MATTER OF THE APPLICATION OF ) PUBLIC SERVICE COMPANY OF COLORADO ) DOCKET NO. A-E FOR APPROVAL OF ITS 0 ELECTRIC ) RESOURCE
Ina Ganguli ENG-SCI 103 Final Project May 16, 2007 Exploring Changes in the Labor Market of Health Care Service Workers in Texas and the Rio Grande Valley I. Introduction The shortage of healthcare workers
Economic Forecast OUTPUT AND EMPLOYMENT 7 8 9 1 11 1 13 1 United States Real GDP $ billions (fourth quarter) $1,996 $1,575 $1,5 $1,9 $15, $15,5 $15,97 $16,6 % change over the four quarters 1.9% -.8% -.%.8%.%.%.%
6.0 - Chapter Introduction In this chapter, you will learn to use moving averages to estimate and analyze estimates of contract cost and price. Single Moving Average. If you cannot identify or you cannot
7 December 2009 NERA Analysis of Energy Supplier Margins By Graham Shuttleworth Even though wholesale energy prices have fallen recently, gas and electricity suppliers are earning very little margin on
Outline: Demand Forecasting Given the limited background from the surveys and that Chapter 7 in the book is complex, we will cover less material. The role of forecasting in the chain Characteristics of
Mathematical Model of Income Tax Revenue on the UK example Svetlana Ivanitskaya, Ilona V. Tregub Financial University under the Government of Russian Federation firstname.lastname@example.org 1. Description of the economic
US and state-level renewable energy adoption rates: 2008-2013 Elena M. Krieger, PhD Physicians, Scientists & Engineers for Healthy Energy January 2014 1 Introduction While the United States power sector
Statistical Sales Forecasting using SAP BPC Capgemini s unique statistical sales forecasting solution integrated with SAP BPC 10.0 helps global fortune 1000 company built robust & accurate sales forecasting
Interrupted time series (ITS) analyses Table of Contents Introduction... 2 Retrieving data from printed ITS graphs... 3 Organising data... 3 Analysing data (using SPSS/PASW Statistics)... 6 Interpreting
Energy Forecasting Methods Presented by: Douglas J. Gotham State Utility Forecasting Group Energy Center Purdue University Presented to: Indiana Utility Regulatory Commission Indiana Office of the Utility
State Farm Bank, F.S.B. 2015 Annual Stress Test Disclosure Dodd-Frank Act Company Run Stress Test Results Supervisory Severely Adverse Scenario June 25, 2015 1 Regulatory Requirement The 2015 Annual Stress
EDUCATION AND VOCABULARY MULTIPLE REGRESSION IN ACTION EDUCATION AND VOCABULARY 5-10 hours of input weekly is enough to pick up a new language (Schiff & Myers, 1988). Dutch children spend 5.5 hours/day
Market Potential and Sales Forecasting There s an old saying derived from a Danish proverb that goes, It s difficult to make predictions, especially about the future. As difficult as predicting the future
APPENDIX 15 Review of demand and energy forecasting methodologies Frontier Economics Energex regulatory proposal October 2014 Assessment of Energex s energy consumption and system demand forecasting procedures
4. Simple regression QBUS6840 Predictive Analytics https://www.otexts.org/fpp/4 Outline The simple linear model Least squares estimation Forecasting with regression Non-linear functional forms Regression
Questioni di Economia e Finanza (Occasional Papers) The evolution of bad debt in Italy during the global financial crisis and the sovereign debt crisis: a counterfactual analysis Number September 216 by
Ministry of Finance Chief Economist - Research, State Revenue and International Affairs June 2013 Forecasts of Macroeconomic Developments, State Revenues from Taxes and Revenue from Other Sources, 2013-2014
Implied Volatility Skews in the Foreign Exchange Market Empirical Evidence from JPY and GBP: 1997-2002 The Leonard N. Stern School of Business Glucksman Institute for Research in Securities Markets Faculty
Measuring the Impact of Tax and Expenditure Limits on Public School Finance in Colorado Executive Summary Most people think that Tax and Expenditure Limits (TELs) keep taxes and government spending lower
Location matters. 3 techniques to incorporate geo-spatial effects in one's predictive model Xavier Conort email@example.com Motivation Location matters! Observed value at one location is