The winners of the second round of the European statistics awards on nowcasting
We are very pleased to announce the winners of the accuracy and reproducibility awards of the Energy Statistics Nowcasting Competitions.
Since 2022, Eurostat has conducted engaging competitions in the fields of nowcasting and web intelligence, with the primary objective of unveiling innovative methodologies and valuable data resources to enhance the production of European statistics. The second round of nowcasting competitions has now concluded, while the third round which will conclude in 2025 is still underway.
We thank all participants for having contributed to the success of this second round through their submissions.
Congratulations to all the winners!
Nowcasting – 2nd round – Energy Statistics Nowcasting Competition
The European Statistics Awards Programme on Nowcasting aims at improving timelines of European Statistics which is a recurring demand by policymakers and other users of European statistics.
We were looking for new approaches based on advanced modelling (possibly using alternative, almost real-time, information) that could give us accurate estimates of key indicators much faster than before.
The second round of the Nowcasting awards was divided into 3 separate competitions, each focusing on the nowcasting of one energy indicator.
Oil and petroleum product deliveries – OIL Competition
Oil and petroleum product deliveries - monthly data [NRG_CB_OILM]
- [O4671] Gas oil and diesel oil
- [GID_CAL] Gross inland deliveries – calculated
- [THS_T] Thousand tons
NRG_CB_OILM belongs to the monthly European statistics that cover the most important energy commodities. For oil and petroleum product deliveries, the data are collected by the reporting countries via separate dedicated questionnaires and subsequently aggregated and transferred to Eurostat. Data on monthly oil delivery figures are collected via standard questionnaires according to Annex C of the Regulation (EC) No 1099/2008 of the European Parliament and of the Council of 22 October 2008 on energy statistics.
Inland gas consumption – GAS Competition
Inland gas consumption - monthly data [NRG_CB_GASM]
- [G3000] Natural gas
- [IC_CAL_MG] Inland consumption of gas - calculated as defined in MOS GAS (Monthly Statistics GAS)
- [TJ_GCV] Terajoule (gross calorific value - GCV)
The NRG_CB_GASM belongs to the monthly European statistics that cover the most important energy commodities. For inland gas consumption, the data are collected by the reporting countries via separate dedicated questionnaires and subsequently aggregated and transferred to Eurostat. Data on monthly inland gas consumption are collected via standard questionnaires according to Annex C of the Regulation (EC) No 1099/2008 of the European Parliament and of the Council of 22 October 2008 on energy statistics.
Electricity availability – ELECTRICITY Competition
Electricity availability - monthly data [NRG_CB_EM]
- [E7000] Electricity
- [AIM] Available to internal market
- [GWH] Gigawatt-hour
NRG-CB_EM belongs to the monthly European statistics that cover the most important energy commodities. For electricity availability, the data are collected by the reporting countries via separate dedicated questionnaires and subsequently aggregated and transferred to Eurostat. Data on monthly electricity availability are collected via standard questionnaires according to Annex C of the Regulation (EC) No 1099/2008 of the European Parliament and of the Council of 22 October 2008 on energy statistics.
As part of the competition requirements, teams were expected to submit at least 6 point estimates of monthly time series in a period of ten months across a minimum of five countries. The submission of estimates for all ten months (and more than five countries) enhanced the team's chances of securing the top position for the accuracy award. Teams competing for the Reproducibility Award were required to provide comprehensive methodological descriptions, including source code.
The Energy Statistics Nowcasting Competition was launched in June 2023, with a final deadline for submissions in March 2024 and documentation in April 2024.
Participation was quite wide as a total of 77 teams, comprised of 107 individuals from 23 countries, signed up for this competition. The results of the evaluation are announced below.
The participants were competing for two types of awards:
- Accuracy Award prizes were given to the top three teams whose entries yielded point estimates closest to the published indicator.
- Reproducibility Award prize was given to the team whose entry shows great potential to be scaled up to European statistics production. The Reproducibility Award was intended to support the best described and documented solutions, with the most innovative, open approach using open data.
Accuracy Award winners - OIL Competition
| Place | Prize | Team name | Team members | Country | Winning entry |
|---|---|---|---|---|---|
| 1st place | 3 000 EUR | outcasted_outliers |
Gergely Attila Kiss Beata Horvath Gabor Lovics Mária Pécs |
Hungary Hungary Hungary Hungary |
Entry 1 |
| 2nd place | 2 000 EUR | YXAnalys | Yingfu Xie | Sweden | Entry 1 |
| 3rd place | 1 000 EUR | Ristretto |
Antoine Palazzolo Thomas Faria Laurent Benichou |
France France France |
Entry 4 |
Accuracy Award winners - GAS Competition
| Place | Prize | Team name | Team members | Country | Winning entry |
|---|---|---|---|---|---|
| 1st place | 3 000 EUR | PM10 |
Daniel Strenger Maximilian Ofner |
Austria Austria |
Entry 1 |
| 2nd place | 2 000 EUR | WattTheForecast | Rafael Finck | Germany | Entry 2 |
| 3rd place | 1 000 EUR | SEER-Group |
Alex Gibberd Andrea D’Orazio Barbara Guardabascio Filippo Moauro Luke Mosley |
UK Italy Italy Italy UK |
Entry 1 |
Accuracy Award winners - ELECTRICITY Competition
| Place | Prize | Team name | Team members | Country | Winning entry |
|---|---|---|---|---|---|
| 1st place | 3 000 EUR | MSME | Jakub Cery | Slovakia | Entry 1 |
| 2nd place | 2 000 EUR | YXAnalys | Yingfu Xie | Sweden | Entry 1 |
| 3rd place | 1 000 EUR | WattTheForecast | Rafael Finck | Germany | Entry 4 |
Reproducibility Award winners
| Competition | Prize | Team name | Team members | Country | Winning entry |
|---|---|---|---|---|---|
| OIL | 5 000 EUR | Ristretto |
Antoine Palazzolo Thomas Faria Laurent Benichou |
France France France |
Entry 4 |
| GAS | 5 000 EUR | PM10 |
Daniel Strenger Maximilian Ofner |
Austria Austria |
Entry 1 |
| ELECTRICITY | 5 000 EUR | WattTheForecast | Rafael Finck | Germany | Entry 4 |
Solutions of the Reproducibility Award winners
Since the Reproducibility award was intended to support the most thoroughly described and documented solutions, with the most innovative, open approach, we are happy to share with you the solutions which won this prize in three competitions (OIL, GAS and ELECTRICITY):
OIL Competition
Team Ristretto’s solution which won the Reproducibility award for presenting the best solution in providing OIL statistics is available on the following link:
GAS Competition
Team PM10’s solution which won the Reproducibility award for presenting the best solution in providing GAS statistics is available on the following link:
ELECTRICITY Competition
Team WattTheForecast’s solution which won the Reproducibility award for presenting the best solution in providing ELECTRICITY statistics is available on the following link:
Final Accuracy Rankings
OIL Competition
| Rank | Team name | Entry 1 score | Entry 2 score | Entry 3 score | Entry 4 score | Entry 5 score | Best entry score |
|---|---|---|---|---|---|---|---|
| 1 | outcasted_outliers | 0.042027 | 0.042027 | ||||
| 2 | YXAnalys | 0.051018 | 0.109153 | 0.144089 | 0.051018 | ||
| 3 | Ristretto | 0.054429 | 0.075745 | 0.065649 | 0.053020 | 0.218081 | 0.053020 |
| 4 | bernese | 0.056685 | 0.119125 | 0.085882 | 0.076056 | 0.116143 | 0.056685 |
| 5 | mb | 0.072188 | 0.057374 | 0.106714 | 0.130079 | 0.135986 | 0.057374 |
| 6 | Koala | 0.057520 | 0.057520 | ||||
| 7 | NowcastAway | 0.104303 | 0.085754 | 0.096864 | 0.107191 | 0.114071 | 0.085754 |
| 8 | Art-energy | 0.111117 | 0.122910 | 0.101101 | 0.094824 | 0.094824 | |
| 9 | dhopp1 | 0.140410 | 0.138225 | 0.096354 | 0.162501 | 0.135397 | 0.096354 |
| 10 | Meteobit | 0.546068 | 0.228308 | 0.227799 | 0.144196 | 0.202519 | 0.144196 |
| 11 | SEER-Group | 0.172231 | 0.171939 | 0.157188 | 0.147108 | 0.147108 | |
| - | Data-Analysis-Trainees | These teams did not provide the same min 5 countries for any of the 6 months, or have some values where MSRE is greater than the set threshold = 0.15 in a reference period. | |||||
| - | Modil | ||||||
| - | SD_CK | ||||||
| - | StatMod23 | ||||||
| - | VarStatic | ||||||
| - | vishleshak | ||||||
GAS Competition
| Rank | Team name | Entry 1 score | Entry 2 score | Entry 3 score | Entry 4 score | Entry 5 score | Best entry score |
|---|---|---|---|---|---|---|---|
| 1 | PM10 | 0.004884 | 0.008114 | 0.007300 | 0.006482 | 0.004884 | |
| 2 | WattTheForecast | 0.147196 | 0.040135 | 0.067719 | 0.043637 | 0.040135 | |
| 3 | SEER-Group | 0.084596 | 0.126408 | 0.089591 | 0.088542 | 0.084596 | |
| 4 | bernese | 0.106243 | 0.204257 | 0.115995 | 0.101341 | 0.148455 | 0.101341 |
| 5 | outcasted_outliers | 0.103326 | 0.103326 | ||||
| 6 | Ristretto | 0.104150 | 0.286616 | 0.123809 | 0.216555 | 0.752225 | 0.104150 |
| 7 | Art-energy | 0.172918 | 0.369266 | 0.120372 | 0.405973 | 0.332392 | 0.120372 |
| 8 | dhopp1 | 0.152287 | 0.142377 | 0.126370 | 0.162102 | 0.126514 | 0.126370 |
| 9 | NowcastAway | 0.174468 | 0.165188 | 0.179158 | 0.175515 | 0.128294 | 0.128294 |
| 10 | Koala | 0.137517 | 0.137517 | ||||
| 11 | YXAnalys | 0.155495 | 0.228429 | 0.345148 | 0.155495 | ||
| 12 | mb | 0.268082 | 0.248891 | 0.251979 | 0.269099 | 0.255028 | 0.248891 |
| 13 | SD_CK | 0.705455 | 0.694704 | 0.694704 | |||
| 14 | Meteobit | 0.942309 | 0.794070 | 0.699056 | 0.728776 | 0.908826 | 0.699056 |
| - | StatMod23 | These teams did not provide the same min 5 countries for any of the 6 months, or have some values where MSRE is greater than the set threshold = 0.15 in a reference period. | |||||
ELECTRICITY Competition
| Rank | Team name | Entry 1 score | Entry 2 score | Entry 3 score | Entry 4 score | Entry 5 score | Best entry score |
|---|---|---|---|---|---|---|---|
| 1 | MSME | 0.004899 | 0.004899 | ||||
| 2 | YXAnalys | 0.005356 | 0.007887 | 0.022491 | 0.005356 | ||
| 3 | WattTheForecast | 0.008559 | 0.007578 | 0.007480 | 0.005548 | 0.005548 | |
| 4 | bernese | 0.010353 | 0.014011 | 0.011341 | 0.012159 | 0.015719 | 0.010353 |
| 5 | Ristretto | 0.011987 | 0.024741 | 0.015855 | 0.023825 | 0.081816 | 0.011987 |
| 6 | Art-energy | 0.019141 | 0.013998 | 0.012334 | 0.018748 | 0.012334 | |
| 7 | Koala | 0.012343 | 0.012343 | ||||
| 8 | mb | 0.012897 | 0.013688 | 0.103223 | 0.077387 | 0.090694 | 0.012897 |
| 9 | outcasted_outliers | 0.016408 | 0.016408 | ||||
| 10 | NowcastAway | 0.062249 | 0.017354 | 0.020504 | 0.017892 | 0.035085 | 0.017354 |
| 11 | EURECOM | 0.017827 | 0.040547 | 0.017827 | |||
| 12 | dhopp1 | 0.024939 | 0.018562 | 0.021341 | 0.025208 | 0.021479 | 0.018562 |
| 13 | SEER-Group | 0.023705 | 0.028577 | 0.023705 | |||
| 14 | SD_CK | 0.045456 | 0.043603 | 0.160923 | 0.204099 | 0.043603 | |
| 15 | StatMod23 | 0.154817 | 0.553351 | 0.154817 | |||
| 16 | Meteobit | 0.206957 | 0.169333 | 0.232109 | 0.169333 | ||
| - | Alberello | These teams did not provide the same min 5 countries for any of the 6 months, or have some values where MSRE is greater than the set threshold = 0.15 in a reference period. | |||||
| - | Data-Analysis-Trainees | ||||||
| - | Modil | ||||||
| - | VarStatic | ||||||
| - | vishleshak | ||||||