GEFCom2014 Load Forecasting Data

The load forecasting track of GEFCom2014 was about probabilistic load forecasting. We asked the contestants to provide one-month ahead hourly probabilistic forecasts on a rolling basis for 15 rounds. In the first round, we provided 69 months of hourly load data and 117 months of hourly temperature data. Incremental load and temperature data was provided in each of the future rounds.

Where to download the data?

The complete data was published as the appendix of our GEFCom2014 paper. If you don't have access to Science Direct, you can downloaded from my Dropbox link HERE. Regardless where you get the data, you should cite this paper to acknowledge the source:

• Tao Hong, Pierre Pinson, Shu Fan, Hamidreza Zareipour, Alberto Troccoli and Rob J. Hyndman, "Probabilistic energy forecasting: Global Energy Forecasting Competition 2014 and beyond", International Journal of Forecasting, vol.32, no.3, pp 896-913, July-September, 2016.

What's in the package?

Unzip the file, you will see the folder "GEFCom2014 Data", which includes five zip files. The data for the probabilistic load forecasting track of GFECom2014 is in the file "GEFCom2014-L_V2.zip". Unzip it, you will see the folder "load", which includes an "Instructions.txt" file and 15 other subfolders. In each folder named as "Task n", there are two files, Ln-train.csv and Ln-benchmark.csv. The train file, together with the train files released in previous rounds, can be used to generate forecasts. The benchmark file includes the forecast generated from the benchmark method.

How to use the data?

Apparently the most straightforward way of using this dataset is to replicate the competition setup and compare results directly with the top entries. Because the data published through GEFCom2014 is quite long (totally 7 years of matching load and temperature data), we can also use this dataset to test methods and models for short term load forecasting.

GEFCom2014-E data

After GEFCom2014, I organized an in-class probabilistic load forecasting competition in Fall 2015 that was open to external participants. My in-class competition setup was very similar to that of GEFCom2014, so I denoted the data for this in-class load forecasting competition as GEFCom2014-E, where E is the abbreviation of "extended". In total, this dataset covers 11 years of hourly temperature and 9 years of hourly load. A top team Florian Ziel was invited to contribute a paper to IJF (see HERE). The readers may replicate the same competition setup and compare results with Ziel's.

Caution

Note that the data I used for GEFCom2014-E was created using ISO New England data. If you want to validate a method using two independent sources, you should not use GEFCom2014-E together with ISO New England data.

Back to Datasets for Energy Forecasting.

Datasets for Energy Forecasting

Reproducible research is a key to advancing knowledge. In energy forecasting, it is necessary and crucial that researchers compare their models and methods using the same datasets. Five years ago when we founded the IEEE Working Group on Energy Forecasting, "lack of benchmark data pool" was one of the issues we identified. Fortunately, things have been changing toward the right direction over the past few years. More and more datasets are being made available to and recognized by the energy forecasting community.

This post will serve as the starting point of a blog series on datasets. In each post, I will feature a dataset and discuss how to use it. I will also host the datasets on Dropbox and provide the links in these posts. Meanwhile, I would like to take a crowd-sourcing approach to making a comprehensive and widely accessible data pool:

• If you can host the datasets through other channels, please contact me. 
• If you know of some public datasets that are not on my list, please contact me. 
• If you have some private datasets that can be made available to the energy forecasting community, please contact me. 

Here is a list of 9 posts with the publicly available data that I have used in my papers. I will update the list with links and additional data sources, so check this page from time to time to see if there is something you need.

Electric load forecasting

1. GEFCom2012
2. GEFCom2014
3. ISO New England
4. RWE npower forecasting challenge 2015

Gas load forecasting

1. RWE npower forecasting challenge 2015

Electricity price forecasting

1. GEFCom2014

Wind power forecasting

1. GEFCom2012
2. GEFCom2014

Solar power forecasting

1. GEFCom2014

Stay tuned...

IJF Special Section on Probabilistic Energy Forecasting: GEFCom2014 Papers and More

As of this week, 21 of the 22 papers for the IJF Special Section on Probabilistic Energy Forecasting are on ScienceDirect (link to the CFP). Many thanks to the GEFCom2014 organizers, participants, and the expert reviewers, whose time and effort warranted an exceptionally high quality collection of energy forecasting papers. Although these papers are not yet pagerized, I can't wait to compile and post this list.

Editorial and GEFCom2014 Introduction Article

• T. Hong, P. Pinson, S. Fan, H. Zareipour, A. Troccoli and R. J. Hyndman, "Probabilistic energy forecasting: Global Energy Forecasting Competition 2014 and beyond" (the competition data is also available on ScienceDirect).

Review Article

• T. Hong and S. Fan, "Probabilistic electric load forecasting: A tutorial review"

Research Articles (Non-GEFCom2014)

• A. Antoniadis, X. Brossat, J. Cugliari and J. Poggi, "A prediction interval for a function-valued forecast model: Application to load forecasting"
• H. Akouemo and R. Povinelli, "Probabilistic anomaly detection in natural gas time series data"
• K. Maciejowska, J. Nowotarski and R. Weron, "Probabilistic forecasting of electricity spot prices using Factor Quantile Regression Averaging"
• A. Bello, J. Reneses, A. Muñoz and A. Delgadillo , "Probabilistic forecasting of hourly electricity prices in the medium-term using spatial interpolation techniques"
• E. Iversen, J. Morales, J. Møller and H. Madsen, "Short-term probabilistic forecasting of wind speed using stochastic differential equations"
• J. Jeon and J. Taylor, "Short-term density forecasting of wave energy using ARMA-GARCH models and kernel density estimation"

Research Articles (GEFCom2014)

• V. Dordonnat, A. Pichavant and A. Pierrot "GEFCom2014 probabilistic Electric load forecasting using time series and semi-parametric regression models"
• J. Xie and T. Hong, "GEFCom2014 probabilistic electric load forecasting: An integrated solution with forecast combination and residual simulation"
• S. Haben and G. Giasemidis, "A hybrid model of kernel density estimation and quantile regression for GEFCom2014 probabilistic load forecasting"
• E. Mangalova and O. Shesterneva, "Sequence of nonparametric models for GEFCom2014 probabilistic electric load forecasting"
• Florian Ziel and Bidong Liu, "Lasso estimation for GEFCom2014 probabilistic electric load forecasting"
• P. Gaillard, Y. Goude and R. Nedellec, "Additive models and robust aggregation for GEFCom2014 probabilistic electric load and electricity price forecasting"
• K. Maciejowska and J. Nowotarski, "A hybrid model for GEFCom2014 probabilistic electricity price forecasting"
• G. Dudek, "Multilayer perceptron for GEFCom2014 probabilistic electricity price forecasting"
• Mark Landry, Thomas P. Erlinger, David Patschke and Craig Varrichio, "Probabilistic gradient boosting machines for GEFCom2014 wind forecasting"
• E. Mangalova and O. Shesterneva, "K-nearest neighbors for GEFCom2014 probabilistic wind power forecasting"
• Y. Zhang and J. Wang, "K-nearest neighbors and kernel density estimator for GEFCom2014 probabilistic wind power forecasting"
• J. Huang and M. Perry, "A semi-empirical approach using gradient boosting and k-nearest neighbors regression for GEFCom2014 probabilistic solar power forecasting"
• G. Nagy, G. Barta, S. Kazi, G. Borbély and G. Simon, "GEFCom2014: Probabilistic solar and wind power forecasting using a generalized additive tree ensemble approach"
• R. Juban, H. Ohlsson, M. Maasoumy, L. Poirier and J. Z. Kolter,, "A multiple quantile regression approach to the wind, solar, and price tracks of GEFCom2014"

Enjoy reading and stay tuned for the next GEFCom!

Probabilistic Energy Forecasting: Global Energy Forecasting Competition 2014 and Beyond

It is really hard to write an introduction for this paper, because there are too many things to highlight the outcome of hundreds of hours invested by the organizers of GEFCom2014 and thousands of hours spent by the GEFCom2014 contestants. In one sentence,

This is a MUST READ paper in energy forecasting. 

In this paper, we

• Summarized 7 papers collected through the special issue CFP process, including one on anomaly detection for gas demand data, two on load forecasting, two on price forecasting, one on wave energy forecasting, and one on wind speed forecasting;
• Introduced the GEFCom2014 together with the in-class probabilistic load forecasting competition I organized in 2015;
• Commented on the methodologies used by the winning teams of GEFCom2014;
• Published the 120MB data used in the four tracks of GEFCom2014 and the in-class competition;
• Made 12 predictions for the next decade of energy forecasting

We sincerely hope that you enjoy reading this paper and can help contribute to the energy forecasting community.

Citation

Tao Hong, Pierre Pinson, Shu Fan, Hamidreza Zareipour, Alberto Troccoli and Rob J. Hyndman, "Probabilistic energy forecasting: Global Energy Forecasting Competition 2014 and beyond", International Journal of Forecasting, in press. working paper available from http://www.drhongtao.com/articles.

Probabilistic Energy Forecasting: Global Energy Forecasting Competition 2014 and Beyond

Tao Hong, Pierre Pinson, Shu Fan, Hamidreza Zareipour, Alberto Troccoli, and Rob J Hyndman

Abstract

The energy industry has been going through a significant modernization process over the last decade. Its infrastructure is being upgraded rapidly. The supply, demand and prices are becoming more volatile and less predictable than ever before. Even its business model is being challenged fundamentally. In this competitive and dynamic environment, many decision-making processes rely on probabilistic forecasts to quantify the uncertain future. Although most of the papers in the energy forecasting literature focus on point or single-valued forecasts, the research interest in probabilistic energy forecasting research has taken off rapidly in recent years. In this paper, we summarize the recent research progress on probabilistic energy forecasting. A major portion of the paper is devoted to introducing the Global Energy Forecasting Competition 2014 (GEFCom2014), a probabilistic energy forecasting competition with four tracks on load, price, wind and solar forecasting, which attracted 581 participants from 61 countries. We conclude the paper with 12 predictions for the next decade of energy forecasting. 

GEFCom2014 Probabilistic Electric Load Forecasting: An Integrated Solution with Forecast Combination and Residual Simulation

Guest Blogger: Jingrui Xie

My adventure on load forecasting started in 2012, when I was the primary developer for the SAS Energy Forecasting solution. Our first customer was North Carolina Electric Membership Cooperation, who used the probabilistic load forecasts generated from our solution for long-term power supply planning.  That project was documented in my first TSG paper co-authored with Dr. Tao Hong and Jason Wilson, which was titled Long term probabilistic load forecasting and normalization with hourly information. Later on, the method proposed in that TSG paper was further investigated by analyzing the residuals. The findings were summarized in our recently published TSG paper On normality assumption in residual simulation for probabilistic load forecasting.

Winners Announced for Global Energy Forecasting Competition 2014

At 2015 PES General Meeting held in Denver, CO, we announced the winning universities and teams of Global Energy Forecasting Competition 2014 (GEFCom2014). The picture below was taken right before the award reception.

I'm delighted to post the winners here.

GEFCom2014 Presentations at PESGM2015

Over the past few months, I have received numerous requests asking for the GEFCom2014 data, papers and presentation files. Recently I managed to compile all the GEFCom2014 Presentations at PESGM2015. Here is the LINK to the ZIP file on OneDrive. I will compile and upload the data by the end of the year. The papers will be published in 2016.

Stay tuned...

What's New in Energy Forecasting - Jun 2015

It's time for the mid-year report and summary of the exciting events in the energy forecasting community:

1. Global Energy Forecasting Competition 2014

The competition was launched in August 2014 and ended in December 2014. We are now in the stage of post-competition activities, such as organizing conference presentations and paper publications. Many finalists of the competition will gather at the IEEE PES General Meeting 2015. The papers are expected to be published in the IJF special issue on probabilistic energy forecasting in early 2016. To follow the update of this competition and future events, please join this LinkedIn group.

2. Activities at IEEE PES General Meeting

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.

Please note that the Wednesday afternoon's panel session is in Plaza 3. We also added a reception on Wednesday evening.

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

Mark Your 2015 Calendar: Tao's Recommended Conferences for Energy Forecasters

Update (3/1/2015): Due to high demand of my load forecasting courses, we added an offering at the New York City in May. 

Update (2/18/2015): I just confirmed two more speakers for ISF2015. Now we have six presentations in two sessions. 

Recently I received many inquiries about recommended energy forecasting conferences in 2015. First of all, I have never attended a conference that is perfectly designed and organized for energy forecasters, which motivates me to organize the ultimate energy forecasting conference. If you have to wait for this ultimate one, close this page and stay tuned for another two years. Otherwise, keep reading. I will provide a list of 6 venues for you to consider, in the chronological order.

1. Tao's load forecasting courses (May 27-29, 2015, New York, NY)

I have taught the fundamental course 15 times. More than 150 energy forecasters have attended the course. (See some statistics based on the first 10 offerings) Recently I have developed a one-day advanced level course for those who want some in depth coverage of the subject and hands on experience of SAS. The links to the courses are listed below:

• Electric Load Forecasting: Fundamentals and Best Practices
• Electric Load Forecasting: Advanced Topics and Case Studies

2. 35th International Symposium on Forecasting (ISF2015, June 21-24, 2015, Riverside, CA)

ISF2015 is a great conference if you want to learn the frontiers of forecasting. I'm organizing an energy forecasting session at ISF2015. The two sessions includes six talks with a balanced mix of state-of-the-art research and practice.

Session Title: Frontiers in Electricity Demand Forecasting I: The State of The Practice
Chair: Tao Hong (University of North Carolina at Charlotte, USA)

• SAS Energy Forecasting: Hourly load forecasting for all horizons
• Bradley Lawson (SAS, USA)
• Combining sister load forecasts
• Tao Hong (University of North Carolina at Charlotte, USA)
• Bidong Liu (University of North Carolina at Charlotte, USA)
• MEFM: An R package for long-term probabilistic forecasting of electricity demand
• Rob J. Hyndman (Monash University, Australia)

Session Title: Frontiers in Electricity Demand Forecasting II: Probabilistic Electric Load Forecasting
Chair: Tao Hong (University of North Carolina at Charlotte, USA)

• Quantile regression algorithms for forecasting uncertainty in electricity smart meters data
• Souhaib Ben Taieb, King Abdullah University of Science and Technology, Saudi Arabia.
• Rob J. Hyndman, Monash University, Australia
• Marc G. Genton, King Abdullah University of Science and Technology, Saudi Arabia.
• Electricity demand interval forecasting with Quantile Regression Averaging
• Jakub Notowarski (Wrocław University of Technology, Poland)
• The myths of residual simulation for probabilistic load forecasting 
• Jingrui Xie (University of North Carolina at Charlotte, USA)

3. 3rd International Conference Energy & Meteorology (ICEM2015, June 22-26, 2015, Denver, CO)

Weather drives electricity demand and wind/solar power generation. The conference has a unique focus on the interdisciplinary field of energy and meteorology. For more information, please read Make a Difference in the Energy & Meteorology World written by the guest blogger and conference chair Alberto Troccoli.

4. Modern Electric Power Systems Conference 2015 (MEPS2015, July 6-9, 2015, Wroclaw, Poland)

If you are in the area of load and price forecasting, you must be familiar with Rafal Weron and his book "Modeling and Forecasting Electricity Loads and Prices: A Statistical Approach" and the recent IJF open access paper "Electricity Price Forecasting: A Review of the State-of-the-art with a Look into the Future". Rafal is based in Wroclaw, Poland. He is also a key player on the organizing committee of MEPS2015. I'm sure he will present some nice forecasting work at the conference.

5. IEEE Power and Energy Society General Meeting 2015 (PESGM2015, July 26-30, 2015, Denver, CO)

We have organized two days of agenda on energy forecasting. A full-day tutorial on "Energy Forecasting in the Smart Grid Era", and a full-day session of GEFCom2014 finalist presentations. I will write another blog post when the schedule is finalized.

6. AEIC/WLRA annual conference (time & location TBD)

This is a joint conference organized by AEIC Load Research Committee and Western Load Research Association. The two groups used to organize conferences separately. Last year was the first time they held a joint conference. I hope they will continue the joint conference this fall.

Mark your calendar and enjoy the trips!

Forecasting and Data Mining

The main difference between forecasting and data mining is on the goal of the task. The goal of forecasting is to make statements about the future, while the goal of data mining is to extract patterns from large datasets. (The term "data mining" was a buzzword 15 years ago to broadly refer to working on the data, which is a misuse.) Many techniques can be applied to both forecasting and data mining, such as artificial neural networks, regression analysis, and clustering analysis, and so forth.

Who's #1? Rating, Ranking and Provisional Leaderboard of GEFCom2014

The 12 evaluation weeks of GEFCom2014 just went by. Many contestants are curious to know about their rankings after such a marathon-type forecasting competition. Last weekend, I created a provisional leaderboard based on the scores I have documented on Inside Leaderboard. In this post, I will share this provisional leaderboard together with the rating and ranking methodology.

Again, this is not the final leaderboard, pending corrections of individual scores (if there were errors) and adjustment of rankings based on the final reports. 

Documentation in Load Forecasting: 4 Reasons and 8 Elements

In load forecasting, especially long term load forecasting, documentation is probably the most important task. The ultimate test of documentation quality is whether the forecasting system has been described in detail so that other people with relevant education background and experience can reproduce the forecasts.

While forecasting is like an adventure, exploring an unknown trail, documentation is like walking the same trail again and again to record what happened in detail. Documentation often requires significant amount of efforts, sometimes more than forecasting itself.

Inside Leaderboard

Update 11/22/2014:

Include a column showing the teams ineligible for the final leaderboard due to missing more than three submissions beating benchmark. 

The first evaluation week (Task 4) of GEFCom2014 just went by. CrowdAnalytix publishes the leaderboard based on the best score of each team in real time. Since some teams made multiple submissions, that leaderboard won't reflect the real positions. To enhance the transparency of our scoring process, I manually pulled the submission log of each track to come up with a more realistic leaderboard. Please understand that:

Towards Winning GEFCom2014 - Six Must Read Recommendations before Evaluation Period Starts

GEFCom2014 had a super strong start. Within the first 4 weeks, we have had 225, 138, 126 and 135 solvers in the load, price, wind and solar tracks respectively. Totally 345 people have joined the LinkedIn group Global Energy Forecasting Competition. While many winning teams of GEFCom2012 came back to GEFCom2014 with very strong performance, several new faces also topped some of the tasks. Having been monitoring the competition from the back end, I am so excited about the ups and downs on the leaderboards. I really wish I could join the game in person.

The evaluation period of GEFCom2014 is starting in less than 2 days. I'd like to offer some recommendations, so that the contestants fully understand what makes a winning solution. Some of them may be overlapping with my previous post, GEFCom2014 is ON - 8 Tips before You Join the Game, but I think it's important to cover them again here.

GEFCom2014 is ON - 8 Tips before You Join the Game

After one full year of planning and implementation, I'm pleased to announce that the Global Energy Forecasting Competition 2014 is ON. Please visit CrowdANALYTIX.com to look for the four tracks.

For now, I will defer all the thanks to the end of the competition. (BTW, it will have to be a long thank-you letter, because so many people have devoted so many days and nights to set up this competition.) Instead, I would spend my last sleepless night before the competition to provide a few tips and instructions to the GEFCom2014 contestants:

10 Recommended Papers for GEFCom2014 Contestants

Update 9/20/2014:

Rafal Weron's review on price forecasting is available on Science Direct with open access. The probabilistic load forecasting review paper written by Shu Fan and myself is currently under review by IJF. The link to the working paper is listed below.

Update 8/14/2014:

Thanks to Rob Hyndman, who generously put Rafal Weron's forthcoming IJF paper on the web. Now we have a super well-written review paper on price forecasting on the list as Ref [11]. This paper will be published in the coming issue of International Journal of Forecasting.

The Global Energy Forecasting Competition 2014 (GEFCom2014) is the first probabilistic forecasting competition in the power and energy industry. As of today, over 200 people from more than 40 countries have signed up the interest list. To help the folks quickly get use to the theme of this competition, we organizers collaboratively picked up 10 recommended papers for GEFCom2014 contestants.

Ref [1-3] are general readings on probabilistic forecasting, energy forecasting and the previous competition respectively. Ref [4-6], Ref [7-8] and Ref [9-10] are for load, wind and solar forecasting respectively. We did not recommend any specific paper on probabilistic price forecasting. The contestants on the price forecasting track may refer to the recommended papers on load and wind forecasting. The literature on probabilistic wind power forecasting is dominantly more extensive and mature than the other three categories. Therefore, we would highly recommend the contestants to read Ref [7-8] regardless which track to work on. Shu and I are preparing the tutorial review on probabilistic load forecasting (Ref [4]), which will serve as a general guideline as well.

What's New in Energy Forecasting - Jun 2014

There are four exciting events going on in the energy forecasting community:

1. Global Energy Forecasting Competition 2014
We will launch GEFCom2014 in August. This competition will feature four tracks: electric load forecasting, electricity price forecasting, wind power forecasting and solar power forecasting. All tracks will be under the theme of probabilistic energy forecasting. The competition will last for 15 weeks with incremental data update every week for rolling forecast. The competition forum is on LinkedIn. For more information, please visit www.gefcom.org.

2. Activities at 2014 IEEE PES General Meeting
IEEE Working Group on Energy Forecasting is hosting a one-day tutorial and a panel session at PESGM2014:

• Energy Forecasting in the Smart Grid Era (tutorial) Sunday, 27 July, 2014 8:00 AM-5:00 PM Chesapeake G
• Load Forecasting: the State of the Practice (panel session) Monday, 28 July, 2014 2:00 PM-5:00 PM National Harbor 4

3. Special Issue on Probabilistic Energy Forecasting | International Journal of Forecasting
Although the abstract submission deadline has passed (refer to Call For Papers), we welcome submissions of high quality work on this topic. In addition, the winners of GEFCom2014 will be invited to submit their methods to this special issue.

4. International Symposium on Forecasting
International Institute of Forecasters is hosting the 34th International Symposium on Forecasting in Rotterdam, The Netherlands, 6/29-7/2, 2014. There are four sessions on energy forecasting: Probabilistic Energy Forecasting; Electricity Demand I; Electricity Demand II; and Oil Prices. The abstracts of the talks are available in the program book. The full conference proceeding with presentations and papers will be available on the conference website. Stay tuned.

A Preview of Global Energy Forecasting Competition 2014

The news article below is what I wrote for The Oracle published by the International Institute of Forecasters. It will show up in the March 2014 issue.

Energy forecasting, in a broad sense, covers a wide range of forecasting problems in the energy industry, such as demand forecasting, generation forecasting, price forecasting and so on. These problems are very attractive to the forecasting community due to the characteristics including high resolution data, multiple seasonality, requirements of low errors and societal necessity.

Announcing Institute Prize for Global Energy Forecasting Competition 2014

To encourage students and faculty participation, we will recognize up to three academic institutes for the excellent performance in the Global Energy Forecasting Competition 2014 (GEFCom2014).

Each individual team that beats the benchmark and ranks top 8 of a track will receive 10 - r points, where r is the ranking of the team. The institute score is the sum of points from all teams associated with the institute. The top three institutes will be recognized with award plaques. The one with the highest score will be awarded the institute price of $500 for the best institute performance. The primary faculty overseeing the participating teams will claim the prize and/or recognition on behalf of a winning institute.

The following terms and conditions apply:

• A team participating and ranking top 8 in multiple tracks will receive the points equal to the sum of points from each track.
• A winning institute has to have at least two teams beating the benchmark(s) and at least one top 8 team. 
• The primary faculty should register all team members prior to the registration deadline. 
• If two institutes receive the same scores, the one with more top 8 teams will have a higher ranking. The priority then goes to the one with more teams beating the benchmark(s).
• If no institute fulfills the above requirements, GEFCom2014 will not award the institute prize. 
• The GEFCom2014 executive committee reserve the rights to interpret and revise the competition rules and other related information.