Monday, June 18, 2018

Combining Probabilistic Load Forecasts

We often find simple averaging as a plausible solution for combining point forecasts. Combining probabilistic forecasts is not that trivial. The literature of combining probabilistic load forecasts is rather limited. Previously, we developed a Quantile Regression Averaging (QRA) method to generate probabilistic load forecasts by combining point forecasts. This work is a follow up, where we combine probabilistic load forecasts to generate a more accurate probabilistic forecast. The method we proposed here is a Constrained Quantile Regression Averaging (CQRA) method, where the parameters of a quantile regression model are non-negative and sum up to 1. We applied the method to loads at both high voltage level and household level, showing better results than the benchmarks.

Among my papers published so far, this one has the shortest title.

Yi Wang, Ning Zhang, Yushi Tan, Tao Hong, Daniel Kirschen, and Chongqing Kang, "Combining probabilistic load forecasts," IEEE Transactions on Smart Grid, in press, available online. (arXiv; IEEE Xplore).

Combining Probabilistic Load Forecasts

Yi Wang, Ning Zhang, Yushi Tan, Tao Hong, Daniel Kirschen, and Chongqing Kang


Probabilistic load forecasts provide comprehensive information about future load uncertainties. In recent years, many methodologies and techniques have been proposed for probabilistic load forecasting. Forecast combination, a widely recognized best practice in point forecasting literature, has never been formally adopted to combine probabilistic load forecasts. This paper proposes a constrained quantile regression averaging (CQRA) method to create an improved ensemble from several individual probabilistic forecasts. We formulate the CQRA parameter estimation problem as a linear program with the objective of minimizing the pinball loss and the constraints that the parameters are nonnegative and summing up to one. We demonstrate the effectiveness of the proposed method using two publicly available datasets, the ISO New England data and Irish smart meter data. Comparing with the best individual probabilistic forecast, the ensemble can reduce the pinball score by 4.39% on average. The proposed ensemble also demonstrates superior performance over nine other benchmark ensembles.

Thursday, June 14, 2018

A Semi-heterogeneous Approach to Combining Crude Oil Price Forecasts

Forecast combination is an effective method to enhance the accuracy. Most combination methods in the literature can be grouped two categories, heterogeneous combination and homogeneous combination, with each having pros and cons. I collaborated with my former visiting scholar Dr. Jue Wang and her colleagues to develop a semi-heterogeneous approach to combining forecasts. We leveraged the decomposition-reconstruction concept, mixing and matching 4 decomposition methods with 4 forecasting techniques. In total this process generates 16 forecasts for combination, which is easier than applying 16 completely different techniques (a.k.a. heterogeneous combination) and more robust than producing 16 different forecasts from one technique (a.k.a. homogeneous combination). Furthermore, the proposed method leads to more accurate forecasts than its counterparts.

Jue Wang, Xiang Li, Tao Hong, and Shouyang Wang, "A semi-heterogeneous approach to combining crude oil price forecasts," Information Sciences, vol.460-461, pp 279-292, September 2018. (ScienceDirect)

A Semi-heterogeneous Approach to Combining Crude Oil Price Forecasts

Jue Wang, Xiang Li, Tao Hong, and Shouyang Wang


Crude oil price forecasting has received increased attentions due to its significant role in the global economy. Accurate crude oil price forecasts often lead to a rapid new production development with higher quality and less cost. Making such accurate forecasts, however, is challenging due to the intrinsic complexity of oil market mechanism. Many techniques have been tested in the crude oil price forecasting literature. Although forecast combination is a well-known method to improve forecast accuracy, generating forecasts using various techniques tend to be labor intensive. How to efficiently generate many individual forecasts for combination becomes a research question in crude oil price forecasting. Recently, several signal decomposition methods have been suggested for processing the oil price signals. In this paper, we propose a semi-heterogeneous approach to combining crude oil price forecasts, which interacts a set of decomposition methods with a set of forecasting techniques. We first decompose the original price series using four decomposition methods, such as Wavelet Analysis, Singular Spectral Analysis, Empirical Mode Decomposition, and Variational Mode Decomposition. We then use four different forecasting techniques, such as Autoregressive Models, Autoregressive Integrated Moving Average Models, Artificial Neural Networks, and Support Vector Regression Models, to forecast the components from each decomposition methods. Finally, we reconstruct the price forecasts from the forecasted components. This process generates 16 price forecasts in total for combination. We test the combination based on all individual forecasts, as well as a subset of the individual forecasts selected using Tabu Search. The experimental results demonstrate that the forecasting models with the addition of a decomposition technique can have an error reduction of 30.6% compared to benchmark models on average. The combined forecasts outperform the individual forecasts on average. Furthermore, comparing with the heterogeneous combination of 4 individual forecasts, the semi-heterogeneous combinations reduce the errors by 56.6% (w/o Tabu Search) and 61.6% (w/ Tabu Search).

Friday, April 27, 2018

Weather Data for Energy Analytics

Being an energy forecaster, I am genuinely interested in meteorology. I even recruited a master student who was a practicing meteorologist in Hawaii (see the blog post about Ying Chen). The more energy forecasting projects I conduct, the more I appreciate the value of weather data. In GEFCom2014, the top 1 place of the solar track was a team of meteorologists from Australia, who completely dominated the track. In GEFCom2017, the top 1 place of the final match was a team of meteorologists from Japan. I truly believe that the energy forecasting community can better leverage meteorology than what we do today. Here is an article about two use cases of weather data for energy analytics. In fact we merged two papers into one by removing the sophisticated mathematics and statistics to keep the story readable to a broad audience. The IEEE Power and Energy Society is so kind to offer the open access to this paper, so that people can read it for free.


Jonathan Black, Alex Hofmann, Tao Hong, Joseph Roberts, and Pu Wang, "Weather data for energy analytics: from modeling outages and reliability indices to simulating distributed photovoltaic fleets," IEEE Power and Energy Magazine, vol.16, no.3, pp 43-53, May-June 2018. (Open AccessIEEE Xplore)

Weather Data for Energy Analytics

From Modeling Outages and Reliability Indices to Simulating Distributed Photovoltaic Fleets

Jonathan Black, Alex Hofmann, Tao Hong, Joseph Roberts, and Pu Wang


Weather impacts virtually all facets of our daily life. As a result, many business sectors are affected by weather conditions, and the power industry is no exception. Weather is a major influencer on system reliability and a key driver of both power supply and demand. In this article, we will demonstrate novel uses of weather data for energy analytics via two utility applications. We first use easily accessible weather data together with regression analysis to model distribution outages and construct a probabilistic view of reliability indices that helps reveal a utility’s reliability trend. We then use high-resolution, commercial-grade weather data to develop realistic simulations of anticipated behind-the-meter photovoltaic (PV) fleets

Friday, April 20, 2018

Training Energy Data Scientists

Traditional power engineering curriculum has been heavily focusing on the engineering aspects of power systems, such as power flow, state estimation, stability and control. Data science has never been a focus in the past. I saw that gap 5 years ago, predicted the shortage of data scientists in the power industry, and left a great place to work to come back to academia. Nowadays, when other business sectors are offering 6-figure salaries to fresh graduates, utilities are having a hard time to compete on the analytics talents. Recently I had the opportunity to collaborate with Prof. David Wenzhong Gao from University of Denver and three other utility executives to put our thoughts in a paper.


Tao Hong, David Wernzhong Gao, Tom Laing, Dale Kruchten, and Jorge Calzada, "Training energy data scientists: universities and industry need to work together to bridge the talent gap," IEEE Power and Energy Magazine, vol.16, no.3, pp 66-73, May-June 2018. (IEEE Xplore)

Training Energy Data Scientists 

Universities and Industry Need to Work Together to Bridge the Talent Gap

Tao Hong, David Gao, Tom Laing, Dale Kruchten, and Jorge Calzada


The workforce crisis is nothing new to the U.S. power industry. It has been a growing concern of both governments and industry organizations since the early 2000s. Meanwhile, the growth of data during the past decade has led to a demand surge for data analytics across all business sectors. The shortage of an electricity workforce and the increasing demand for data analytics present an emerging challenge as well as opportunity for university power engineering programs to bridge the data analytics talent gap. After gathering various perspectives from members of academia, industry, and government, we propose an interdisciplinary and entrepreneurial approach to revising the traditional power engineering curriculum for training the next generation of energy data scientists.

Saturday, April 14, 2018

Call For Papers: Forecasting for Social Good | International Journal of Forecasting

International Journal of Forecasting

Special Issue on Forecasting for Social Good

The area of forecasting and its link to decision making has been under research for decades. Whilst there have been many influential contributions seeking to examine the effects of forecasting under financial and economic motives, very little has been contributed (both in regular conferences and journal publications) on forecasting with social impact – that is forecasting for the social good, regardless of the financial implications, or optimizations attempted based on economic terms.

The International Journal of Forecasting (IJF) is excited to announce this Call For Papers for the special issue on “Forecasting for Social Good”. The purpose of the special issue is to attract high quality papers that are concerned with the social impact of forecasting.

Areas of interest include, but are not limited to:
  • Health and healthcare
  • Humanitarian operations
  • Disaster relief
  • Education
  • Social services
  • Environment
  • Sustainability
  • Sharing economy
  • Transportation
  • Urban planning
  • Fraud, collusion, and corruption
  • Government policy
  • Poverty
  • Privacy
  • Cyber security
  • Crime and terrorism

Submission Deadline: 31 October 2018.

Submission Guidelines:

To submit a paper for consideration for the Special Issue, please upload your paper online and include a cover letter clearly indicating that the paper is for the special issue “Forecasting for Social Good”. The webpage for online submission is Instructions for authors are provided at . All papers will follow IJF’s double-blind refereeing process. For further information about the Special Issue, please contact the guest editors.

Guest Editors

Bahman Rostami-Tabar, Cardiff University, UK

Michael Porter, University of Alabama, USA

Tao Hong, University of North Carolina at Charlotte, USA

Friday, April 13, 2018

Which Model is the Best?

Recently Rob Hyndman blogged about the history of forecasting competitions. I have read the post three times already. I learned something new each time. I wish every reader of this blog can also read that post and learn something from the history. Nevertheless, I would like to highlight a paragraph here:
[...] This reveals a view commonly held (even today) that there is some single model that describes the data generating process, and that the job of a forecaster is to find it. This seems patently absurd to me — real data comes from much more complicated, non-linear, non-stationary processes than any model we might dream up — and George Box famously dismissed it saying “All models are wrong but some are useful”.
I have been in the field load forecasting for a little more than 10 years. During the past decade, I received many "the best" questions: Which model is the best? Which technique is the best? Which software is the best? Which variable is the best? ...

There are some variants to these questions: Do you use neural networks? Do you use population in long term load forecasting? Do you use normal weather? Do you think demand response can reduce peak demand? Do you think a MAPE value of 20% is too high? ...

While the folks who asked these questions may expect a crisp answer of the best model, technique, software, variable, or a YES/NO answer, I had to disappoint them with "depends", sometimes followed by a lengthy elaboration.
It depends on the data, the business needs, the production environment, and many other factors... 
I don't have a single model to sell to my clients as "the best model". I recommend proper methodologies to the clients after making a comprehensive evaluation of their situations.

I think the root of these "the best" questions is that commonly held view:
There is some single model that describes the data generating process. 
I disagree with this view, and don't know where the view is originally coming from. I wish someone can write another history article to explain the source.

In load forecasting, there is no universally best model. We need empirical studies, many empirical studies, to show some evidence that one method is superior in some aspect. That's why I have been promoting reproducible research and benchmarking data pool and models.

In computational complexity and optimization, there is actually a theorem, no free lunch theorem

Sunday, March 25, 2018

Call For Papers: Data Analytics for Energy, Water, and Environment | IEEE Transactions on Engineering Management

IEEE Transactions on Engineering Management

Special Issue on Data Analytics for Energy, Water, and Environment

In 2008, the U.S. National Academy of Engineering proposed 14 grand challenges for engineering, of which five were related to energy, water, and environment. The world population has exceeded 7.6 billion in 2018, up from 6.7 billion in 2008. A common goal of the human beings is to meet the global need of energy and water without sacrificing the environment.

A growing interest in the recent decade is data analytics. Many industries have embarked analytics to extract actionable insights from the data and to make informed business decisions. This special issue of the IEEE Transactions on Engineering Management is aimed at highlighting the state of the art in data analytics applications in energy, water and environment.

The topics of interest include but are not limited to:
  • Quantitative models for water-energy nexus
  • Evaluating energy technology portfolios
  • Data integration and infusion for energy, water and environment applications
  • Cyber-physical systems for sustainable development
  • Power supply and air pollution
  • Building energy efficiency and indoor air quality
  • Energy efficiency analytics enabled by smart meter data
  • Clean energy/water access 
  • Energy systems planning under climate and water challenges
  • Risks and uncertainties in energy engineering projects
  • Energy management considering consumer behaviors
  • Sustainable supply chain design and inventory control
  • Water and energy management in sustainable agriculture
  • Environment, energy, and alternative transportation technologies

Submission Process

Please prepare the manuscript according to IEEE-TEM’s guidelines and submit it the journal’s Manuscript Central site. Please clearly state in the cover letter that the submission is for this special issue. 


Interested authors send 2-page extended abstracts to the Corresponding Guest Editor by August 31st 2018
Decisions on acceptance of abstracts by November 30th 2018
Full papers submitted by May 31st 2019

Corresponding Guest Editor

Dr. Tao Hong (
Department of Systems Engineering and Engineering Management 
University of North Carolina at Charlotte

Guest Editors

Dr. Hua Cai, Purdue University
Dr. Gang He, Stony Brook University
Dr. Guiping Hu, Iowa State University
Dr. Yueming Qiu, University of Maryland
Dr. Ekundayo Shittu, George Washington University
Dr. Hongliang Zhang, Louisiana State University
Dr. Ning Zhang, Tsinghua University


Dr. Tugrul Daim, Portland State University

Friday, March 23, 2018

Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges

About 10 years ago, the term "smart grid" was officially defined in the Energy Independence and Security Act of 2007 (EISA-2007). Soon after that, many power companies started their smart meter deployments. As of 2016, more than 70 million smart meters were installed in the united states. The world installed base was projected to reach 780 million by 2020, pushed by the mass roll-outs in China. We are now sitting on a gold mine of data collected by these smart meters. Last year I gave a forecast:
The energy companies will be moving more Gigabytes of data than GWh of electricity.
The scientific community has been trying to understand the smart meter data and get some actionable insights out of it. Thousands of papers have been published in the recent decade on the various aspects of smart meter data analytics. Last year, I worked with my collaborators in Tsinghua University to complete a review of smart meter data analytics. The paper was just put on the IEEE Xplore yesterday.

This is the longest paper ever published by the IEEE Transactions on Smart Grid. I'm sure reading this 24-page review article can save the readers significant amount of time from digging thousands of papers in the literature. Load forecasters may find some interesting stuff in Section III, which is dedicated to load forecasting in the smart grid era.


Yi Wang, Qixin Chen, Tao Hong, and Chongqing Kang, "Review of smart meter data analytics: applications, methodologies, and challenges," IEEE Transactions on Smart Grid, in press. (working paper; IEEE Xplore)

Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges

Yi Wang, Qixin Chen, Tao Hong, and Chongqing Kang


The widespread popularity of smart meters enables an immense amount of fine-grained electricity consumption data to be collected. Meanwhile, the deregulation of the power industry, particularly on the delivery side, has continuously been moving forward worldwide. How to employ massive smart meter data to promote and enhance the efficiency and sustainability of the power grid is a pressing issue. To date, substantial works have been conducted on smart meter data analytics. To provide a comprehensive overview of the current research and to identify challenges for future research, this paper conducts an application-oriented review of smart meter data analytics. Following the three stages of analytics, namely, descriptive, predictive and prescriptive analytics, we identify the key application areas as load analysis, load forecasting, and load management. We also review the techniques and methodologies adopted or developed to address each application. In addition, we also discuss some research trends, such as big data issues, novel machine learning technologies, new business models, the transition of energy systems, and data privacy and security.

Saturday, March 3, 2018

A Look Back at ISEA2017

About a year ago, I posted 7 reasons to attend ISEA2017. Last June, we had our very first International Symposium on Energy Analytics at the beautiful city Cairns, Australia. 50 researchers and practitioners from industry and academia across 16 countries attended the symposium. 36 of them also attended ISF2017.

The two-day agenda included 5 keynote presentations, 13 contributed presentations selected from 31 abstract submissions, and 5 GEFCom2017 finalists presentations. At the end of the first day, June 22, 2017, we hosted the GEFCom2017 award ceremony followed by a two-hour cocktail reception. The program agenda, presentations and pictures can be downloaded via THIS LINK.

We have been processing the full paper submissions since then. So far, three papers are currently in press with the Journal of Modern Power Systems and Clean Energy, one is in press with the Power & Energy Magazine. The remaining papers are going through the peer review system of the International Journal of Forecasting. The accepted ones together with the GEFCom2017 articles are going to be published in an energy forecasting special section edited by Pierre Pinson and myself.  

I want to thank many people who made ISEA2017 a successful event, the keynote speakers, the presenters, the GEFCom2017 winners, and all the attendees of ISEA2017. Special thanks to Rob Hyndman and George Athanasopoulos for the strong local support, our photographer Mitchell O'Hara-Wild for capturing the happy moments, and Pam Stroud for handling the registrations and helping keep things in order. 

Last but not least, big thanks to our sponsors, International Institute of Forecasters, Tangent Works and Journal of Modern Power Systems and Clean Energy, for their generous financial support that helped to keep the registration fees low. 

Many people have asked me about the next ISEA. I know there will be the next one, but I don't know when and where. If you have any idea or want to help with organizing ISEA, let me know!