Intraday price forecast

More and more flexible consumers and power and storages are actively participating in the power exchange in order to save on electricity costs or to generate revenues. In addition to Day-Ahead Trading, trading on the Intraday Market has also increased significantly in recent years. Intraday trading is more volatile than the Day-Ahead Market due to shorter time periods and in some cases shorter products. On the one hand, this increases the risk in marketing, but on the other hand, it can also result in attractive profit opportunities.

For trading on the power exchange, the market participants need price forecasts on the basis of which they can make their trading decisions. A short-term forecast period is often decisive for the marketing of flexibilities. For this purpose, we have developed a fundamental model for the short-term price forecast of the quarter-hourly intraday auction. This is based on hourly Day-Ahead Prices, as well as load and generation forecasts.

We also forecast the price uncertainty of continuous intraday prices depending on the intraday auction based on historical data.

Context: Products and trading hours on the power exchange

The majority of the trading volume on the European power exchange EPEX consists of the Day-Ahead Trade, which takes place at 12 noon on the day before delivery. Products traded are hourly products and block products of several hours. Increasingly, however, the trading volume in the subsequent and thus shorter-term Intraday Trading is also increasing. In Germany, Intraday trading begins with the Intraday Opening Auction at 3 p.m. the day before, in which power products are traded in quarter-hourly resolution. Quarter-hourly products are also traded in the subsequent continuous Intraday Trading. Unlike the Day-Ahead Market and the Intraday Auction, prices are not formed here as a volume-maximizing unit price, but matching bids are matched directly and bidders pay or receive the price they bid (pay-as-bid).

Forecast of the Intraday-Auction Price

The price forecast of the Intraday Auction Prices is formed between the hourly Day-Ahead Auction and the Intraday Auction (GER). This is based in particular on the hourly Day-Ahead Price, as well as the load and renewable forecasts at the time of the Day-Ahead Auction. These data are publicly available at [1]. The calculation is based on a regression analysis in which the deviations of the Intraday Price from the Day-Ahead Price are predicted on the basis of the load and renewable forecasts. For the forecast quality of the resulting price forecast, the period from which data is used is particularly decisive. While a too short time period does not provide sufficient data for a reliable forecast, a too long forecast period cannot sufficiently reflect current changes on the market. This is especially important at times when, as is currently the case, electricity prices are subject to considerable uncertainty. The methodology can be used equivalently for forecasting continuous Intraday Prices, but the accuracy decreases compared to the Intraday Auction due to the larger time gap to Day-Ahead trading.


Price uncertainty between intraday auction and continuous trading

The price uncertainty of continuous intraday trading is specified via the ID1 price index of EPEX, which is formed from the bids of the last hour before delivery. For this purpose, an interval is formed based on the historical standard deviation of the difference between ID1 index and intraday auction price, in which the ID1 index is expected to occur to almost 68% [2]. Analogous to the prediction of intraday price characteristics, the forecast period is also of great importance for the prediction of price uncertainties. The deviation of continuous prices from the intraday auction price also exhibit situational uncertainties, which can be differentiated by the level of residual load (and hence electricity prices). In the current market environment, high residual loads and thus high price levels tend to result in greater uncertainties in the continuous intraday trading.

Further Information: