Jakub Nowotarski

An empirical comparison of alternative schemes for combining electricity spot price forecasts

In this paper we investigate the use of forecast averaging for electricity spot prices. While there is an increasing body of literature on the use of forecast combinations, there is only a small number of applications of these techniques in the area of electricity markets. In this comprehensive empirical study we apply seven averaging and one selection scheme and perform a backtesting analysis on day-ahead electricity prices in three major European and US markets. Our findings support the additional benefit of combining forecasts for deriving more accurate predictions, however, the performance is not uniform across the considered markets. Interestingly, equally weighted pooling of forecasts emerges as a viable robust alternative compared with other schemes that rely on estimated combination weights. Overall, we provide empirical evidence that also for the extremely volatile electricity markets, it is beneficial to combine forecasts from various models for the prediction of day-ahead electricity prices. In addition, we empirically demonstrate that not all forecast combination schemes are recommended.

Probabilistic Load Forecasting via Quantile Regression Averaging on Sister Forecasts

The majority of the load forecasting literature has been on point forecasting, which provides the expected value for each step throughout the forecast horizon. In the smart grid era, the electricity demand is more active and less predictable than ever before. As a result, probabilistic load forecasting, which provides additional information on the variability and uncertainty of future load values, is becoming of great importance to power systems planning and operations. This paper proposes a practical methodology to generate probabilistic load forecasts by performing quantile regression averaging on a set of sister point forecasts. There are two major benefits of the proposed approach. It can leverage the development in the point load forecasting literature over the past several decades and it does not rely so much on high-quality expert forecasts, which are rarely achievable in load forecasting practice. To demonstrate the effectiveness of the proposed approach and make the results reproducible to the load forecasting community, we construct a case study using the publicly available data from the Global Energy Forecasting Competition 2014. Compared with several benchmark methods, the proposed approach leads to dominantly better performance as measured by the pinball loss function and the Winkler score.

Merging quantile regression with forecast averaging to obtain more accurate interval forecasts of Nord Pool spot prices

We evaluate a recently proposed method for constructing prediction intervals, which utilizes the concept of quantile regression (QR) and a pool of point forecasts of different time series models. We find that in terms of interval forecasting of Nord Pool day-ahead prices the new QR-based approach significantly outperforms prediction intervals obtained from standard, as well as, semi-parametric autoregressive time series models.

Modeling and forecasting of the long-term seasonal component of the EEX and Nord Pool spot prices

We present the results of a study on modeling and forecasting of the long-term seasonal component (LTSC) of electricity spot prices. We consider a vast array of models including linear regressions, monthly dummies, sinusoidal decompositions and wavelet smoothers. We find that in terms of forecasting EEX and Nord Pool spot prices up to a year ahead, wavelet-based models significantly outperform all considered piecewise constant and sine-based models. This result challenges the traditional approach to deseasonalize spot electricity prices by fitting monthly dummies or sinusoidal functions. We also find that extending the calibration window up to four years does not improve the results; two- and especially three-year windows lead to better spot price forecasts.

Real-time ultrasound-guided infraclavicular axillary vein cannulation: A prospective study in mechanically ventilated critically ill patients

PURPOSE The main purpose of this study was to define the venipuncture and catheterization success rates and early mechanical complication rates of ultrasound-guided infraclavicular axillary vein cannulation. MATERIALS AND METHODS We performed in-plane, real-time, ultrasound-guided infraclavicular axillary vein catheterizations under emergency and nonemergency conditions in mechanically ventilated, critically ill patients. RESULTS We performed 202 cannulation attempts. One hundred and twenty-six procedures (62.4%) were performed under emergency conditions. The puncture of the axillary vein was successful in 98.5% of patients, and the entire procedure success rate was 95.1% (95% confidence interval, 91.1%-97.6%). For the majority of patients (84.1%; P<.001, exact test), the venipuncture occurred during the first attempt. We noted a 22.4% overall complication rate, and most of the complications were malpositions (13.4%). We observed 8.5% of cases with potentially serious complications (puncture of the axillary artery and needle contact with the brachial plexus) and 1 case (0.5%) of pneumothorax. The puncture of the axillary artery occurred in 5 (2.5%) patients. CONCLUSIONS In-plane, real-time, ultrasound-guided, infraclavicular axillary vein cannulation in mechanically ventilated, critically ill patients is a safe and reliable method of central venous cannulation and can be considered to be a reasonable alternative to other central venous catheterization techniques.

Real-time, ultrasound-guided infraclavicular axillary vein cannulation for renal replacement therapy in the critical care unit—A prospective intervention study

PURPOSE The cannulation of the axillary vein for renal replacement therapy is a rarely performed procedure in the critical care unit. We defined the venipuncture and catheterization success rates and early mechanical complication rates of this technique in critical care patients with acute kidney injury. MATERIALS AND METHODS Twenty-nine mechanically ventilated patients with clinical indications for insertion of temporary hemodialysis catheters enrolled in a registered trial (NCT01919528) as a pilot cohort. We performed 29 real-time, ultrasound-guided infraclavicular axillary vein cannulation attempts for renal replacement therapy. We defined the venipuncture and catheterization success rates and early mechanical complication rates for this technique. RESULTS The puncture of the axillary vein was successful in 28 (96.5%) patients. In 22 patients (75.9%), venipuncture occurred during the first attempt and in 6 patients during the second (20.7%). The overall cannulation success rate was 93.1% (95% confidence interval, 77%-99%). We noted 6.8% potentially serious complications rate, 10.3% minor complications rate, and 0% life-threatening early mechanical complications. We achieved an 89.6% renal replacement therapy success rate and low rate of catheters malfunction. CONCLUSIONS Real-time, ultrasound-guided, infraclavicular axillary vein cannulation for renal replacement therapy in the critical care unit is a reliable method of dual-lumen hemodialysis catheter insertion and can be considered a reasonable alternative to jugular and femoral routes in special clinical circumstances.

A new infraclavicular landmark-based approach to the axillary vein as an alternative method of central venous cannulation

Purpose We developed the new technique of the axillary vein catheterization, which is connected with the determination of only two anatomical points of reference for puncture site identification. The primary outcome of this study was to determine the rate of successful catheterizations and the assessment of procedure success rate, depending on cannulation side as well as physician experience. The secondary objective was to evaluate the early complication rate and to determine whether this method can be used in clinical practice. Methods The methodology of this prospective, cohort study included catheterization of the axillary vein via the infraclavicular approach. All procedures were performed by the first two authors, each of whom had different levels of experience with the technique. The choice of the cannulation side was based on clinical factors, and the technique was identical on the right and left sides. Results The cannulation success rate reached 85.6% (N = 153). The correlation between physician experience and the procedure success rate as well as between cannulation side and procedure success rate were not significant. A common early complication was a puncture of axillary artery (14.4%) with the following proper cannulation in the majority of patients (77.3%, p<0.01, exact test). Conclusions The procedure success rate, 85.6% (95% CI [80.0, 91.2%]), is comparable to other landmark-based techniques of the central vein cannulation. The moderately high percentage of axillary artery puncture points out that the approach should be used only as an alternative method to the central vein catheterization.

Principle component forecast averaging — Application to the UK electricity market

In the paper, a forecast combination via dimension reduction techniques is applied to forecasting electricity spot prices. We propose to use a Factor Averaging (FA) method, which is based on a principle component approach. This methodology allows to extract information from a large panel of forecasts and helps to solve two important issues: collinearity of forecasts and common bias of different predicting methods. The empirical performance of FA and other forecast averaging methods is evaluated using the data describing the UK market. Twenty four individual methods are used to calculate one day ahead forecast of 48 hal-hourly prices. The outcomes prove that FA is more efficient in terms of the RMSE than other forecast combining approaches. Gains from using FA varies between 8-20% when comparing the average daily RMSE and between 1-40% when individual half-hourly periods are considered.