Skip to content

Latest commit

 

History

History
207 lines (151 loc) · 11.9 KB

README.md

File metadata and controls

207 lines (151 loc) · 11.9 KB

emarketcrawlR: crawling data from the european energy market (EPEX SPOT, EEX)

Goal

This R package provides functions to crawl the european energy market EPEX SPOT in Paris at https://www.epexspot.com and https://www.eex.com There are three trading auctions at the EPEX SPOT (status quo: all implemented):

  • Intraday Continuous Trading: getIntradayContinuousEPEXSPOT()
  • Intraday Auction: getIntradayAuctionEPEXSPOT()
  • Day-Ahead Auction: getDayAheadAuctionEPEXSPOT()

For all auctions, except the Intraday Auction (only german), french, german and swiss market data is available. The crawiling functions for this trading auctions return price data (in €/MWh) and volume (MW) as well as different block prices for that day, especially base and peak load prices.

Status quo the PHELIX DE Future price data of the EEX is implemented:

  • Phelix-DE Futures: getPHELIXDEFuturesEEX()

Get Started

Installing

When installing this package you should at least use the R version 3.3.0 (2016-05-03). For the library dependecies see the section below. You can easily install this R package by using the install_github() function from the devtools package:

library(devtools)
install_github("wagnertimo/emarketcrawlR")

Library dependencies

Before using this R package, please check that you have installed the following R packages. Normally with the installation of the package those dependencies will also be installed. If not, you have to do it manually.

  • httr
  • xml2
  • XML
  • lubridate
  • dplyr
  • logging
  • purrr
  • rjson
  • scales
  • grid
  • gridExtra

Usage

Activate the package with library(emarketcrawlR) then continue with:

1. Continuous Intraday Trading at EPEX SPOT

The function getIntradayContinuousEPEXSPOT() retrieves the continuous intraday trading data of the EPEX SPOT in Paris. Therefore it crawls the website https://www.epexspot.com/en/market-data/intradaycontinuous/intraday-table/. You can specify a time period in the format YYYY-MM-DD, a trading product (the time in minutes 60, 30, 15) and the country ("DE", "FR", "CH"). The returned data.frame contains information about the Low(€/MWh), High(€/MWh), Last(€/MWh), Weighted Avg.(€/MWh), Index(€/MWh), ID3(€/MWh, only for German Market), Buy and Sell Volume(MW) as well as the Base and Peak Load(€/MWh).

A big disadvantage of the function is, that the website of EPEX SPOT only provides information of two days on one site. Hence a request to retrieve a longer time period of data can take awhile since the function has to make a request for every two days within that time interval.

# Set Logging to print out the state of process including a progress bar
setLogging(TRUE)

# Get the 15min (default: hour data) trading price data in the given time period of the german cont. intra. at EPEX SPOT
prices <- getIntradayContinuousEPEXSPOT("2017-05-20", "2017-05-26", "15", "DE")

head(prices)
# Output:
#              DateTime  Low High Last Weighted_Avg   Idx   ID3 Buy_Vol Sell_Vol Index_Base Index_Peak
# 1 2017-05-20 00:00:00 17.5 35.0 24.0        25.74 25.74 25.58   329.8    420.8      21.22      18.82
# 2 2017-05-20 00:15:00 -1.2 32.0 32.0        20.48 20.48 20.48   264.5    286.5      21.22      18.82
# 3 2017-05-20 00:30:00 -3.7 33.0 28.0        21.85 21.85 21.85   347.5    347.5      21.22      18.82
# 4 2017-05-20 00:45:00 12.8 32.0 30.0        24.90 24.90 24.91   510.1    510.1      21.22      18.82
# 5 2017-05-20 01:00:00 15.0 31.8 29.5        23.19 23.19 23.17   292.7    292.7      21.22      18.82
# 6 2017-05-20 01:15:00  2.8 25.0 13.7        10.75 10.75 10.69   220.4    220.4      21.22      18.82

Plotting Continuous Intraday Trading data

The package has a built-in function for a quick plot to visualize the price (Low, High, Last in €/MWh) and volume (buy and sell in GWh) data. The function does not return a plot object since it uses grid.arrange. The plot is directly outputted. The plot contains heavy graphic elements. Therfore it is not appropriate to use it for a large time period.

# Get new data. Just a three days
lastPrices <- getIntradayContinuousEPEXSPOT("2017-05-26", "2017-05-28", "15", "DE")

qplotIntradayContinuous(lastPrices)

2. Intraday Auction at EPEX SPOT

The function getIntradayAuctionEPEXSPOT() retrieves the intraday auction data of the EPEX SPOT in Paris. Therefore it crawls the website https://www.epexspot.com/en/market-data/intradayauction/quarter-auction-table/. You can specify a time period in the format YYYY-MM-DD. This kind of auction is only available for the german market. The returned data contains the intraday auction prices (15min base) and its volume as well as the block prices, especially the base and peak prices and volumes.

On the website are the last past six days of the requested date given. This allows to crawl 7 days of data with one request (this is a bit more than in the case of the getIntradayContinuousEPEXSPOT() function). Keep in mind that larger time frames will take awhile.

The blocks are defined as follows:

  • Off-Peak (00-00 & 07-00 & 20-00 & 23-00)
  • Baseload Price with Volume (00-00 & 23-00)
  • Off-Peak 1 (00-00 & 07-00)
  • Peakload Price with Volume (08-00 & 19-00)
  • Sun Peak (10-00 & 15-00)
  • Off-Peak 2 (20-00 & 23-00)
# Set Logging to print out the state of process including a progress bar
setLogging(TRUE)

# Get the 15min auction price data in the given time period of the german intra. auction market at EPEX SPOT. 
auctionPrices <- getIntradayAuctionEPEXSPOT("2017-05-19", "2017-05-26")

head(auctionPrices)
# Output:
#              DateTime Prices Volume OffPeak OffPeak1 SunPeak OffPeak2 BasePrice BaseVolume PeakPrice PeakVolume
# 1 2017-05-19 00:00:00  34.28  694.7   30.42    27.78   37.28     35.7     33.88      53489     37.33    27486.8
# 2 2017-05-19 00:15:00  30.59  291.1   30.42    27.78   37.28     35.7     33.88      53489     37.33    27486.8
# 3 2017-05-19 00:30:00  28.75  276.8   30.42    27.78   37.28     35.7     33.88      53489     37.33    27486.8
# 4 2017-05-19 00:45:00  23.50  546.1   30.42    27.78   37.28     35.7     33.88      53489     37.33    27486.8
# 5 2017-05-19 01:00:00  31.32  528.1   30.42    27.78   37.28     35.7     33.88      53489     37.33    27486.8
# 6 2017-05-19 01:15:00  26.95  185.7   30.42    27.78   37.28     35.7     33.88      53489     37.33    27486.8

3. Day-Ahead-Auction at EPEX SPOT

The function getDayAheadAuctionEPEXSPOT() retrieves the day-ahead auction data of the EPEX SPOT in Paris. Therefore it crawls the website https://www.epexspot.com/en/market-data/dayaheadauction/auction-table/. You can specify a time period in the format YYYY-MM-DD and french ("FR"), german ("DE") or swiss ("CH") market data. The returned data contains the auction prices and its volume as well as the block prices, especially the base and peak prices and volumes.

The blocks are defined as follows:

  • Middle Night (01-04)
  • Early Morning (05-08)
  • Late Morning (09-12)
  • Early Afternoon (13-16)
  • Rush Hour (17-20)
  • Off-Peak 2 (21-24)
  • Baseload (01-24)
  • Peakload (09-20)
  • Night (01-06)
  • Off-Peak 1 (01-08)
  • Business (09-16)
  • Off-Peak (01-08 & 21-24)
  • Morning (07-10)
  • High Noon (11-14)
  • Afternoon (15-18)
  • Evening (19-24)
  • Sun Peak (11-16)
# Set Logging to print out the state of process including a progress bar
setLogging(TRUE)

# Get the hourly day-ahead auction price data in the given time period of the french auction market at EPEX SPOT. 
auction <- getDayAheadAuctionEPEXSPOT("2017-05-26", "2017-05-28", "FR")

head(auction)
# Output:
#              DateTime Prices  Volume MiddleNight EarlyMorning LateMorning EarlyAfternoon RushHour OffPeak2 Night OffPeak1 Business OffPeak
# 1 2017-05-26 00:00:00  28.38 11792.9       25.68        30.65       35.47          28.58    36.05    38.09 32.42    33.37    25.57   28.16
# 2 2017-05-26 01:00:00  25.71 11414.8       25.68        30.65       35.47          28.58    36.05    38.09 32.42    33.37    25.57   28.16
# 3 2017-05-26 02:00:00  24.65 11339.5       25.68        30.65       35.47          28.58    36.05    38.09 32.42    33.37    25.57   28.16
# 4 2017-05-26 03:00:00  23.96 11580.9       25.68        30.65       35.47          28.58    36.05    38.09 32.42    33.37    25.57   28.16
# 5 2017-05-26 04:00:00  24.33 12430.2       25.68        30.65       35.47          28.58    36.05    38.09 32.42    33.37    25.57   28.16
# 6 2017-05-26 05:00:00  26.39 12424.9       25.68        30.65       35.47          28.58    36.05    38.09 32.42    33.37    25.57   28.16
#    Morning HighNoon Afternoon Evening SunPeak BasePrice BaseVolume PeakPrice PeakVolume
# 1   32.02    31.47     36.66   31.62   29.58     32.42   331828.5     33.37   179938.8
# 2   32.02    31.47     36.66   31.62   29.58     32.42   331828.5     33.37   179938.8
# 3   32.02    31.47     36.66   31.62   29.58     32.42   331828.5     33.37   179938.8
# 4   32.02    31.47     36.66   31.62   29.58     32.42   331828.5     33.37   179938.8
# 5   32.02    31.47     36.66   31.62   29.58     32.42   331828.5     33.37   179938.8
# 6   32.02    31.47     36.66   31.62   29.58     32.42   331828.5     33.37   179938.8

4. Phelix-DE Futures at EEX

The function getPHELIXDEFuturesEEX() retrieves the Phelix-DE Futures at the EEX. Therefore it crawls the website https://www.eex.com/en/market-data/power/futures/phelix-de-futures. You can specify a time period in the format YYYY-MM-DD, a trading product (Day, Weekend, Week, Month, Quarter or Year, RECOMMENDED: USE DAY). The returned data.frame mimics the table at the EEX website (+ Best Bid/ Best Ask volume). The prices are given in EUR/MWh and the volume in MWh. The name column contains no Date or DateTime object (it is simply a factor).

The data is retrieved by sequentially scraping the data of each day (observation). Therefore the data of the next day (observation, regarding the input date) is retrieved. The weekend data (saturday and sunday ) is retrieved at once on friday dates. As you can see, the next data observation is retrieved (because tables are showing a lot of null values) and hence the function is optimized for the "Day" product. For other products like Week or Year, you will get only the next week or next year data given the input date. Note also, that the website only provides a limited history of the price data! This also depends on the product (Day, Week etc.)

# Set Logging to print out the state of process including a progress bar
setLogging(TRUE)

# Get the price data of the Day Phelix-DE Futures at EEX from 2017-08-02 till 2017-08-04
prices <- getPHELIXDEFuturesEEX("2017-08-02", "2017-08-04", "Day")

head(prices)
# Output:
#        Name Block Product BestBid BestBidVolume BestAsk BestAskVolume NoContracts LastPrice AbsChange LastTime LastVolume
# 1 2017.08.02  Base     Day       0             0       0             0          75     33.65      0.15    08:55        600
# 2 2017.08.02  Peak     Day       0             0       0             0           0        NA     -0.76     <NA>         NA
# 3 2017.08.03  Base     Day       0             0       0             0          75     27.65     -1.35    10:17        600
# 4 2017.08.03  Peak     Day       0             0       0             0           0        NA     -1.31     <NA>         NA
# 5 2017.08.04  Base     Day       0             0       0             0           0        NA      2.31     <NA>         NA
# 6 2017.08.04  Peak     Day       0             0       0             0           0        NA      0.15     <NA>         NA
#   SettlementPrice VolumeExchange VolumeTradeRegister OpenInterest
# 1           33.05           1800                   0            0
# 2           34.99              0                   0            0
# 3           28.07           1800                   0           25
# 4           28.44              0                   0            0
# 5           24.56              0                   0           75
# 6           21.65              0                   0            0