Wednesday, May 20, 2015

Energy Forecasting Activities at PESGM2015

IEEE Power & Energy Society just released the technical program agenda for PESGM2015. I'm pleased to highlight the two days on energy forecasting organized by our working group.

Saturday, July 26, 2015

Energy Forecasting in the Smart Grid Era (full-day tutorial)
8:00 AM - 5:00 PM
Instructors: Tao Hong, Shu Fan, Hamidreza Zareipour, and Pierre Pinson

Wednesday, July 29, 2015

Global Energy Forecasting Competition 2014: an Overview (panel session)
8:00 AM - 12:00 PM Plaza Court 3
Chair: Tao Hong
  1. Probabilistic Electric Load Forecasting   S. Fan. 
  2. Probabilistic Electricity Price Forecasting   H. Zareipour. 
  3. Probabilistic Wind Power Forecasting   P. Pinson. 
  4. Probabilistic Solar Power Forecasting   A. Troccoli. 
  5. GEFCom2014 Institute Prize - University of North Carolina at Charlotte   T. Hong. 
  6. GEFCom2014 Institute Prize - Tsinghua University   C. Kang. 
  7. Probabilistic Load and Price Forecasting - Tololo   R. Nedellec. 
  8. Probabilistic Price, Wind and Solar Forecasting - C3 Green Team   Z. Kolter. 
  9. Probabilistic Wind and Solar Forecasting - dmlab   G. Nagy. 

Global Energy Forecasting Competition 2014: Finalist Presentations (panel session)
1:00 PM - 5:00 PM Governor's Square 17
Chair: Tao Hong
  1. Probabilistic Electric Load Forecasting - Adada   A. Pierrot. 
  2. Probabilistic Electric Load Forecasting - Jingrui (Rain) Xie   J. Xie. 
  3. Probabilistic Electric Load Forecasting - OxMath   S. Haben. 
  4. Probabilistic Electricity Price Forecasting - Team Poland   J. Nowotarski. 
  5. Probabilistic Electricity Price Forecasting - Pat1   F. Lemke. 
  6. Probabilistic Wind Power Forecasting - kPower   M. Landry. 
  7. Probabilistic Wind Power Forecasting - Yao Zhang   Y. Zhang. 
  8. Probabilistic Solar Power Forecasting - Gang-gang   J. Huang. 
  9. Probabilistic Solar Power Forecasting - UT_Argonne   D. Lee. 

Thursday, May 14, 2015

Probabilistic Load Forecasting via Quantile Regression Averaging on Sister Forecasts

Although the probabilistic load forecasting literature can be traced back to 1970s, the importance of the subject was not well recognized until recent years. There are several approaches to producing probabilistic load forecasts, such as generating weather scenarios to feed to point forecasting models, applying probabilistic modeling and forecasting techniques, and identifying the density function of residuals. This paper starts a whole new category for probabilistic load forecasting methods - combining point load forecasts.

The proposed methodology includes two parts:
  1. Generating sister point forecasts. The concept of sister models/forecasts is a brand new in this field. The models were first developed from my big data paper. This paper further materializes the concept and applies it to probabilistic load forecasting. 
  2. Combining forecasts with quantile regression. While quantile regression is not a rare technique, there has not been many studies that apply quantile regression. In this paper, we use quantile regression to combining the point forecasts. The combination is dominantly better than the probabilistic load forecasts from individual models. 
The paper was accepted by IEEE Transactions on Smart Grid this week. The working paper is available HERE. I will update the citation once the paper is on IEEE Xplore.

Bidong Liu, Jakub Nowotarski, Tao Hong and Rafal Weron, "Probabilistic Load Forecasting via Quantile Regression Averaging on Sister Forecasts", IEEE Transactions on Smart Grid, accepted, working paper available from

Probabilistic Load Forecasting via Quantile Regression Averaging on Sister Forecasts

Bidong Liu, Jakub Nowotarski, Tao Hong and RafaƂ Weron


Majority of the load forecasting literature has been on point forecasting, which provides the expected value for each step throughout the forecast horizon. In the smart grid era, the electricity demand is more active and less predictable than ever before. As a result, probabilistic load forecasting, which provides additional information on the variability and uncertainty of future load values, is becoming of great importance to power systems planning and operations. This paper proposes a practical methodology to generate probabilistic load forecasts by performing Quantile Regression Averaging (QRA) on a set of sister point forecasts. There are two major benefits of the proposed approach: 1) it can leverage the development in the point load forecasting literature over the past several decades; and 2) it does not rely so much on high quality expert forecasts, which are rarely achievable in load forecasting practice. To demonstrate the effectiveness of the proposed approach and make the results reproducible to the load forecasting community, we construct a case study using the publicly available data from the Global Energy Forecasting Competition 2014. Comparing with several benchmark methods, the proposed approach leads to dominantly better performance as measured by the pinball loss function and the Winkler score.  

Tuesday, May 5, 2015

HONG Analytics LLC

I started my career as a consultant when I was pursuing my first MS degree. Then I developed my MS thesis and PhD dissertation, each from a consulting project. Both of them were later commercialized and now being used by many utilities worldwide. Most of my research ideas were inspired by the consulting projects through working closely with the utility analysts, managers and executives. After testing and validating these research ideas through several field implementations, I converted them to teaching materials for my courses. These courses fully packed with solid fundamental knowledge, best industry practices, advanced topics and a wide range of case studies have been very well received by the industry (see what the clients are saying), generating even more consulting business.

As mentioned in why I left a great place to work - from industry to academia, majority of my academic job is made of consulting, teaching and research. Going through such a cycle has been quite rewarding for both my clients and myself. Nevertheless, some of my clients are still complaining about the lengthy and verbose contracting process. To make the cycle even more enjoyable and productive, I decided to incorporate HONG Analytics LLC to formally house my consulting practices outside the university.

HONG Analytics provides premium training and consulting services in the area of energy and retail analytics. While being a full-time professor, I can allocate up to one day per week of my time to HONG Analytics. I would like to take on those projects that are relatively small in scale (i.e., less than $100k) and relatively short in performance period (i.e., less than 6 months). While I'm happy to help with conventional short and long term load forecasting projects, I would give preference to the projects that are challenging and requiring heavy-duty analytics. Since revenue is not the a KPI of my business, I am also willing to provide the services free of charge for those projects with high value to the industry.

For more information, please visit

Tuesday, April 21, 2015

BigDEAL Students Receiving Awards from Lee College of Engineering

Today (April 21, 2015), William States Lee College of Engineering hosted an award luncheon to celebrate student achievements. I'm very pleased to see two of our BigDEAL students being recognized in this event:
Moreover, we had the first full house today with all BigDEAL members gathered on campus. Here is a selfie we took after the award luncheon. Thanks to Bidong's selfie stick :)
BigDEAL Selfie - Spring 2015
Congratulations, Rain and Bidong!

Friday, April 17, 2015

Language for Rejecting Energy Forecasting Papers

As a volunteer in the peer review system, I'm handling (as an editor & editor-in-chief) and reviewing (as a reviewer) at least 10 papers per month. While I try to perform thorough review for each paper following these four steps, the consequence is that I spend more time helping others improve and publish papers than working on my own papers. As the editorial workload keeps growing, I have been trying to figure out a more efficient way of performing editorial services. One phenomenon I have observed is that most papers (about 80% or more) get rejected at the end. Most of my editorial services are for top journals such as TSG and IJF, of which the rejection rate is even higher than 80%. In fact many papers share similar reasons for rejection. I guess one way is to standardize the rejection language, so that I don't have to type the same reasons each time. To save time for other editors and reviewers in the same peer review system, I would be more than happy to share the following rejection language.

Thursday, April 16, 2015

Jiali Liu - The First Alumna of BigDEAL

Today (April 16, 2015), Jiali Liu just defended her MS thesis on combining sister load forecasts, which made her the second graduate of BigDEAL. Since the first BigDEAL graduate Rain is continuing her PhD study here, Jiali becomes the first alumna of BigDEAL.

Jiali received her bachelor degree in International Business and Trade from Wuhan University of Technology, and her second bachelor degree in Business Administration from University of Pittsburgh, both in 2012. In August 2013, Jiali joined the Master of Science in Engineering Management program of UNC Charlotte. In May 2014. She joined BigDEAL to conduct MS thesis research under my supervision. She received her SAS Base Programmer and SAS Advanced Programmer certifications in summer 2014. She was part of the BigDEAL team winning the Analytics2014 poster contest. She was also a key contributor to the EISPC/NARUC Load Forecasting Case Study. She participated in the Global Energy Forecasting Competition 2014 with a top 8 place in the probabilistic wind power forecasting track.

Jiali received her Graduate Certificate in Energy Analytics in December 2014. This May, she will receive her MS in Engineering Management degree with a concentration in Energy Systems. After that, Jiali will move to Raleigh working as a Business Analyst at Evalueserve.

Congratulations, Jiali, and all the best!

Thursday, April 9, 2015

Crystal Ball Lessons in Predictive Analytics

For a long time, I have had the idea of writing an article about "how much benefit are we getting from reducing load forecast errors". A few months ago I got a request from EnergyBiz to contribute an article. I thought this "valuation" topic would be a good fit. So here is Crystal Ball Lessons in Predictive Analytics.

Monday, April 6, 2015

Job Opening at SAS: Solutions Architect

Posting a job for my friends and former colleagues at SAS. SAS is well-known to be a best place to work. As a former SAS employee enjoying the SAS benefits in person, I have to say that whatever you read from the web are underestimating the comfort working for SAS. (BTW, this is why I left SAS.) If you are interested in joining SAS, please apply HERE.

Solutions Architect (Commercial Business Unit - Energy & Utilities)-20006794


Thursday, April 2, 2015

Jingrui Xie - The First Graduate of BigDEAL

Today (April 2, 2015), Jingrui (Rain) Xie just defended her MS thesis on retail energy forecasting, which made her the first graduate of BigDEAL.

Rain received her B.A. in Finance from Sun Yat-sen University in 2009, and her M.A. in Economics from Duke University in 2011. In August 2013, Rain joined the Master of Science in Engineering Management program of UNC Charlotte as my first student. Since then she has published two journal papers, one on long term probabilistic load forecasting, and the other on retail energy forecasting. She was a key contributor to the EISPC/NARUC Load Forecasting Case Study. She was part of the BigDEAL team that topped the NPower Forecasting Challenge 2015.  As one of the top teams of the Global Energy Forecasting Competition 2014, she was invited by International Journal of Forecasting to write a paper describing her winning methodology. This summer, Rain will be attending two conferences, International Symposium on Forecasting and IEEE Power and Energy Society General Meeting, to present her work on probabilistic load forecasting.

Rain will continue her study in the Ph.D. Program in Infrastructure and Environmental Systems (INES) here at UNC Charlotte. Her dissertation topic is on probabilistic load forecasting.

Congratulations, Rain!

Monday, March 30, 2015

Call For Papers: GlobalSIP'15 Symposium on Signal and Information Processing for Optimizing Future Energy Systems

Posting this CFP per request of Dr. Hao Zhu.

GlobalSIP'15 Symposium on 
Signal and Information Processing for Optimizing Future Energy Systems
Orlando, Florida, USA, December 14-16, 2015