Ensemble forecast of photovoltaic power with online CRPS learning

We provide probabilistic forecasts of photovoltaic (PV) production, for several PV plants located in France up to 6 days of lead time, with a 30-min timestep. First, we derive multiple forecasts from numerical weather predictions (ECMWF and Météo France), including ensemble forecasts. Second, our parameter-free online learning technique generates a weighted combination of the production forecasts for each PV plant. The weights are computed sequentially before each forecast using only past information. Our strategy is to minimize the Continuous Ranked Probability Score (CRPS). We show that our technique provides forecast improvements for both deterministic and probabilistic evaluation tools.     

Combining predictive distributions for the statistical post-processing of ensemble forecasts

Statistical post-processing techniques are now used widely for correcting systematic biases and errors in the calibration of ensemble forecasts obtained from multiple runs of numerical weather prediction models. A standard approach is the ensemble model output statistics (EMOS) method, which results in a predictive distribution that is given by a single parametric law, with parameters that depend on the ensemble members. This article assesses the merits of combining multiple EMOS models based on different parametric families. In four case studies with wind speed and precipitation forecasts from two ensemble prediction systems, we investigate the performances of state of the art forecast combination methods and propose a computationally efficient approach for determining linear pool combination weights. We study the performance of forecast combination compared to that of the theoretically superior but cumbersome estimation of a full mixture model, and assess which degree of flexibility of the forecast combination approach yields the best practical results for post-processing applications.     

An ensemble approach to GEFCom2017 probabilistic load forecasting

We present an ensembling approach to medium-term probabilistic load forecasting which ranked second out of 73 competitors in the defined data track of the GEFCom2017 qualifying match. In addition to being accurate, the ensemble method is highly scalable, due to the fact that it had to be applied to nine quantiles in ten zones and for six rounds. Candidate forecasts were generated using random settings for input data, covariates, and learning algorithms. The best candidate forecasts were averaged to create the final forecast, with the number of candidate forecasts being chosen based on their accuracy in similar validation periods.     

Probabilistic access forecasting for improved offshore operations

Improving access is a priority in the offshore wind sector, driven by the opportunity to increase revenues, reduce costs, and improve safety at operational wind farms. This paper describes a novel method for producing probabilistic forecasts of safety-critical access conditions during crew transfers. Methods of generating density forecasts of significant wave height and peak wave period are developed and evaluated. It is found that boosted semi-parametric models outperform those estimated via maximum likelihood, as well as a non-parametric approach. Scenario forecasts of sea-state variables are generated and used as inputs to a data-driven vessel motion model, based on telemetry recorded during 700 crew transfers. This enables the production of probabilistic access forecasts of vessel motion during crew transfer up to 5 days ahead. The above methodology is implemented on a case study at a wind farm off the east coast of the UK.     

Comparing probabilistic forecasts of the daily minimum and maximum temperature

Understanding changes in the frequency, severity, and seasonality of daily temperature extremes is important for public policy decisions regarding heat waves and cold snaps. A heat wave is sometimes defined in terms of both the daily minimum and maximum temperature, which necessitates the generation of forecasts of their joint distribution. In this paper, we develop time series models with the aim of providing insight and producing forecasts of the joint distribution that can challenge the accuracy of forecasts based on ensemble predictions from a numerical weather prediction model. We use ensemble model output statistics to recalibrate the raw ensemble predictions for the marginal distributions, with ensemble copula coupling used to capture the dependency between the marginal distributions. In terms of time series modelling, we consider a bivariate VARMA-MGARCH model. We use daily Spanish data recorded over a 65-year period, and find that, for the 5-year out-of-sample period, the recalibrated ensemble predictions outperform the time series models in terms of forecast accuracy.     

Operational solar forecasting for the real-time market

Despite the significant progress made in solar forecasting over the last decade, most of the proposed models cannot be readily used by independent system operators (ISOs). This article proposes an operational solar forecasting algorithm that is closely aligned with the real-time market (RTM) forecasting requirements of the California ISO (CAISO). The algorithm first uses the North American Mesoscale (NAM) forecast system to generate hourly forecasts for a 5-h period that are issued 12 h before the actual operating hour, satisfying the lead-time requirement. Subsequently, the world’s fastest similarity search algorithm is adopted to downscale the hourly forecasts generated by NAM to a 15-min resolution, satisfying the forecast-resolution requirement. The 5-h-ahead forecasts are repeated every hour, following the actual rolling update rate of CAISO. Both deterministic and probabilistic forecasts generated using the proposed algorithm are empirically evaluated over a period of 2 years at 7 locations in 5 climate zones.     

Testing the predictive accuracy of COVID-19 forecasts

We test the predictive accuracy of forecasts of the number of COVID-19 fatalities produced by several forecasting teams and collected by the United States Centers for Disease Control and Prevention for the epidemic in the United States. We find three main results. First, at the short horizon (1 week ahead) no forecasting team outperforms a simple time-series benchmark. Second, at longer horizons (3 and 4 week ahead) forecasters are more successful and sometimes outperform the benchmark. Third, one of the best performing forecasts is the Ensemble forecast, that combines all available predictions using uniform weights. In view of these results, collecting a wide range of forecasts and combining them in an ensemble forecast may be a superior approach for health authorities, rather than relying on a small number of forecasts.     

Calibration of deterministic NWP forecasts and its impact on verification

Deterministic forecasts (as opposed to ensemble or probabilistic forecasts) issued by numerical weather prediction (NWP) models require post-processing. Such corrective procedure can be viewed as a form of calibration. It is well known that, based on different objective functions, e.g., minimizing the mean square error or the mean absolute error, the calibrated forecasts have different impacts on verification. In this regard, this paper investigates how a calibration directive can affect various aspects of forecast quality outlined in the Murphy–Winkler distribution-oriented verification framework. It is argued that the correlation coefficient is the best measure for the potential performance of NWP forecast verification when linear calibration is involved, because (1) it is not affected by the directive of linear calibration, (2) it can be used to compute the skill score of the linearly calibrated forecasts, and (3) it can avoid the potential deficiency of using squared error to rank forecasts. Since no single error metric can fully represent all aspects of forecast quality, forecasters need to understand the trade-offs between different calibration strategies. To echo the increasing need to bridge atmospheric sciences, renewable energy engineering, and power system engineering, as to move toward the grand goal of carbon neutrality, this paper first provides a brief introduction to solar forecasting, and then revolves its discussion around a solar forecasting case study, such that the readers of this journal can gain further understanding on the subject and thus potentially contribute to it.     

Comparison of intraday probabilistic forecasting of solar irradiance using only endogenous data

Accurate solar forecasts are necessary to improve the integration of solar renewables into the energy grid. In recent years, numerous methods have been developed for predicting the solar irradiance or the output of solar renewables. By definition, a forecast is uncertain. Thus, the models developed predict the mean and the associated uncertainty. Comparisons are therefore necessary and useful for assessing the skill and accuracy of these new methods in the field of solar energy.The aim of this paper is to present a comparison of various models that provide probabilistic forecasts of the solar irradiance within a very strict framework. Indeed, we consider focusing on intraday forecasts, with lead times ranging from 1 to 6 h. The models selected use only endogenous inputs for generating the forecasts. In other words, the only inputs of the models are the past solar irradiance data. In this context, the most common way of generating the forecasts is to combine point forecasting methods with probabilistic approaches in order to provide prediction intervals for the solar irradiance forecasts. For this task, we selected from the literature three point forecasting models (recursive autoregressive and moving average (ARMA), coupled autoregressive and dynamical system (CARDS), and neural network (NN)), and seven methods for assessing the distribution of their error (linear model in quantile regression (LMQR), weighted quantile regression (WQR), quantile regression neural network (QRNN), recursive generalized autoregressive conditional heteroskedasticity (GARCHrls), sieve bootstrap (SB), quantile regression forest (QRF), and gradient boosting decision trees (GBDT)), leading to a comparison of 20 combinations of models.None of the model combinations clearly outperform the others; nevertheless, some trends emerge from the comparison. First, the use of the clear sky index ensures the accuracy of the forecasts. This derived parameter permits time series to be deseasonalized with missing data, and is also a good explanatory variable of the distribution of the forecasting errors. Second, regardless of the point forecasting method used, linear models in quantile regression, weighted quantile regression and gradient boosting decision trees are able to forecast the prediction intervals accurately.     

Bayesian VAR forecasts,survey information,and structural change in the euro area

We incorporate external information extracted from the European Central Bank’s Survey of Professional Forecasters into the predictions of a Bayesian VAR using entropic tilting and soft conditioning. The resulting conditional forecasts significantly improve the plain BVAR point and density forecasts. Importantly, we do not restrict the forecasts at a specific quarterly horizon but their possible paths over several horizons jointly since the survey information comes in the form of one- and two-year-ahead expectations. As well as improving the accuracy of the variable that we target, the spillover effects on “other-than-targeted” variables are relevant in size and are statistically significant. We document that the baseline BVAR exhibits an upward bias for GDP growth after the financial crisis, and our results provide evidence that survey forecasts can help mitigate the effects of structural breaks on the forecasting performance of a popular macroeconometric model.     

Conditional Probabilistic Population Forecasting

Since policy-makers often prefer to think in terms of alternative scenarios, the question has arisen as to whether it is possible to make conditional population forecasts in a probabilistic context. This paper shows that it is both possible and useful to make these forecasts. We do this with two different kinds of examples. The first is the probabilistic analog of deterministic scenario analysis. Conditional probabilistic scenario analysis is essential for policy-makers because it allows them to answer "what if" type questions properly when outcomes are uncertain. The second is a new category that we call "future jump-off date forecasts". Future jump-off date forecasts are valuable because they show policy-makers the likelihood that crucial features of today's forecasts will also be present in forecasts made in the future.     

Evaluating quantile forecasts in the M5 uncertainty competition

Probabilistic forecasts are necessary for robust decisions in the face of uncertainty. The M5 Uncertainty competition required participating teams to forecast nine quantiles for unit sales of various products at various aggregation levels and for different time horizons. This paper evaluates the forecasting performance of the quantile forecasts at different aggregation levels and at different quantile levels. We contrast this with some theoretical predictions, and discuss potential implications and promising future research directions for the practice of probabilistic forecasting.     

Efficient forecast tests for conditional policy forecasts

Central Banks regularly make forecasts, such as the Fed’s Greenbook forecast, that are conditioned on hypothetical paths for the policy interest rate. While there are good public policy reasons to evaluate the quality of such forecasts, up until now, the most common approach has been to ignore their conditional nature and apply standard forecast efficiency tests. In this paper we derive tests for the efficiency of conditional forecasts. Intuitively, these tests involve implicit estimates of the degree to which the conditioning path is counterfactual and the magnitude of the policy feedback over the forecast horizon. We apply the tests to the Greenbook forecast and the Bank of England’s inflation report forecast, finding some evidence of forecast inefficiency. Nonetheless, we argue that the conditional nature of the forecasts made by central banks represents a substantial impediment to the analysis of their quality—stronger assumptions are needed and forecast inefficiency may go undetected for longer than would be the case if central banks were instead to report unconditional forecasts.     

Comparing trained and untrained probabilistic ensemble forecasts of COVID-19 cases and deaths in the United States

The U.S. COVID-19 Forecast Hub aggregates forecasts of the short-term burden of COVID-19 in the United States from many contributing teams. We study methods for building an ensemble that combines forecasts from these teams. These experiments have informed the ensemble methods used by the Hub. To be most useful to policymakers, ensemble forecasts must have stable performance in the presence of two key characteristics of the component forecasts: (1) occasional misalignment with the reported data, and (2) instability in the relative performance of component forecasters over time. Our results indicate that in the presence of these challenges, an untrained and robust approach to ensembling using an equally weighted median of all component forecasts is a good choice to support public health decision-makers. In settings where some contributing forecasters have a stable record of good performance, trained ensembles that give those forecasters higher weight can also be helpful.     

Short term load forecasting and the effect of temperature at the low voltage level

Short term load forecasts will play a key role in the implementation of smart electricity grids. They are required for optimising a wide range of potential network solutions on the low voltage (LV) grid, including the integration of low carbon technologies (such as photovoltaics) and the utilisation of battery storage devices. Despite the need for accurate LV level load forecasts, much of the literature has focused on the individual household or building level using data from smart meters, or on aggregates of such data. This study provides a detailed analysis of several state-of-the-art methods for both point and probabilistic LV load forecasts. We evaluate the out-of-sample forecast accuracies of these methodologies on 100 real LV feeders, for horizons from one to four days ahead. In addition, we also test the effect of the temperature (both actual and forecast) on the accuracy of load forecasts. We present some important results on the drivers of forecasts accuracy as well as on the empirical comparison of point and probabilistic forecast measures.     

Does judgment improve macroeconomic density forecasts?

This paper presents empirical evidence on how judgmental adjustments affect the accuracy of macroeconomic density forecasts. Judgment is defined as the difference between professional forecasters’ densities and the forecast densities from statistical models. Using entropic tilting, we evaluate whether judgments about the mean, variance and skew improve the accuracy of density forecasts for UK output growth and inflation. We find that not all judgmental adjustments help. Judgments about point forecasts tend to improve density forecast accuracy at short horizons and at times of heightened macroeconomic uncertainty. Judgments about the variance hinder at short horizons, but can improve tail risk forecasts at longer horizons. Judgments about skew in general take value away, with gains seen only for longer horizon output growth forecasts when statistical models took longer to learn that downside risks had reduced with the end of the Great Recession. Overall, density forecasts from statistical models prove hard to beat.     

Measuring output gap nowcast uncertainty

We propose a methodology for gauging the uncertainty in output gap nowcasts across a large number of commonly-deployed vector autoregressive (VAR) specifications for inflation and the output gap. Our approach utilises many output gap measures to construct ensemble nowcasts for inflation using a linear opinion pool. The predictive densities for the latent output gap utilise weights based on the ability of each specification to provide accurate probabilistic forecasts of inflation. In an application based on US real-time data, nowcasting over the out-of-sample evaluation period from 1991q2 to 2010q1, we demonstrate that a system of bivariate VARs produces well-calibrated ensemble densities for inflation, in contrast to univariate autoregressive benchmarks. The implied nowcast densities for the output gap are multimodal and indicate a considerable degree of uncertainty. For example, we assess the probability of a negative output gap at around 45% between 2004 and 2007. Despite the Greenspan policy regime, there still remained a substantial risk that the nowcast for output was below potential in real time. We extend our methodology to include distinct output gap measures, based on alternative filters, and show that, in our application, the nowcast density for the output gap is sensitive to the detrending method.     

Evaluating probabilistic forecasts of extremes using continuous ranked probability score distributions

Verifying probabilistic forecasts for extreme events is a highly active research area because popular media and public opinions are naturally focused on extreme events, and biased conclusions are readily made. In this context, classical verification methods tailored for extreme events, such as thresholded and weighted scoring rules, have undesirable properties that cannot be mitigated, and the well-known continuous ranked probability score (CRPS) is no exception.In this paper, we define a formal framework for assessing the behavior of forecast evaluation procedures with respect to extreme events, which we use to demonstrate that assessment based on the expectation of a proper score is not suitable for extremes. Alternatively, we propose studying the properties of the CRPS as a random variable by using extreme value theory to address extreme event verification. An index is introduced to compare calibrated forecasts, which summarizes the ability of probabilistic forecasts for predicting extremes. The strengths and limitations of this method are discussed using both theoretical arguments and simulations.     

One model and various experts: Evaluating Dutch macroeconomic forecasts

The Netherlands Bureau for Economic Policy Analysis (CPB) uses a large macroeconomic model to create forecasts of various important macroeconomic variables. The outcomes of this model are usually filtered by experts, and it is the expert forecasts that are made available to the general public. In this paper we re-create the model forecasts for the period 1997-2008 and compare the expert forecasts with the pure model forecasts. Our key findings from the first time that this unique database has been analyzed are that (i) experts adjust upwards more often; (ii) expert adjustments are not autocorrelated, but their sizes do depend on the value of the model forecast; (iii) the CPB model forecasts are biased for a range of variables, but (iv) at the same time, the associated expert forecasts are more often unbiased; and that (v) expert forecasts are far more accurate than the model forecasts, particularly when the forecast horizon is short. In summary, the final CPB forecasts de-bias the model forecasts and lead to higher accuracies than the initial model forecasts.     

Using weather ensemble predictions in electricity demand forecasting

Weather forecasts are an important input to many electricity demand forecasting models. This study investigates the use of weather ensemble predictions in electricity demand forecasting for lead times from 1 to 10 days ahead. A weather ensemble prediction consists of 51 scenarios for a weather variable. We use these scenarios to produce 51 scenarios for the weather-related component of electricity demand. The results show that the average of the demand scenarios is a more accurate demand forecast than that produced using traditional weather forecasts. We use the distribution of the demand scenarios to estimate the demand forecast uncertainty. This compares favourably with estimates produced using univariate volatility forecasting methods.