Short term load forecasting using a hybrid neural network

Short Term Load Forecasting (STLF) is very important from the power systems grid operation point of view. STLF involves forecasting load demand in a short term time frame. The short term time frame may consist of half hourly prediction up to weekly prediction. Accurate forecasting would benefit the utility in terms of reliability and stability of the grid ensuring adequate supply is present to meet with the load demand. Apart from that it would also affect the financial performance of the utility company. An accurate forecast would result in better savings while maintaining the security of the grid. This paper outlines the STLF using a novel hybrid online learning neural network, known as the Gaussian Regression (GR). This new hybrid neural network is a combination of two existing online learning neural networks which are the Gaussian Adaptive Resonance Theory (GA) and the Generalized Regression Neural Network (GRNN). Both GA and GRNN implemented online learning, but each of them suffers from limitation. Originally GA is used for unsupervised clustering by compressing the training samples into several categories. A supervised version of GA is available, namely Gaussian ARTMAP (GAM). However, the GAM is still not capable on solving regression problem. On the other hand, GRNN is designed for solving real value estimation (regression) problem, but the learning process would involve of memorizing all training samples, hence high computational cost. The hybrid GR is considered an enhanced version of GRNN with compression ability while still maintains online learning properties. Simulation results show that GR has comparable prediction accuracy and has less prototype as compared to the original GRNN as well as the Support Vector Regression.

Incorporating differences in marginal utilities of time across activities in a time allocation model

Most existing activity time allocation models assume that individuals allocate their time to different activities over a period in such a way that the marginal utilities of time across activities are equal. Their argument is that, if not equal, an individual is free to allocate more time to those activities whose marginal utilities of time are higher and, finally allocates the optimal time to each activity with equal marginal utility. However, such an ideal situation may not always prevail in reality, especially when an individual is under income constraint and/or under intense time pressure. In order to incorporate such differences in marginal utilities of time across activities, we enrich the traditional activity time allocation model by explicitly including income constraint and by adding marginal extension activity choice model. As an application, the developed integrated model is used to estimate the value of activity time during weekends in Tokyo. The results are encouraging in that they forecast the individual time allocation more accurately and estimate realistically the value of activity time for each activity in a set of different activities than do by existing traditional models.

Improving load forecasting accuracy through combination of best forecasts

Neural network (NN) models have been widely used in the literature for short-term load forecasting. Their popularity is mainly due to their excellent learning and approximation capability. However, their forecasting performance significantly depends on several factors including initializing parameters, training algorithm, and NN structure. To minimize negative effects of these factors, this paper proposes a practically simple, yet effective and an efficient method to combine forecasts generated by NN models. The proposed method includes three main phases: (i) training NNs with different structures, (ii) selecting best NN models based on their forecasting performance for a validation set, and (iii) combination of forecasts for selected best NNs. Forecast combination is performed through calculating the mean of forecasts generated by best NN models. The performance of the proposed method is examined using real world data set. Comparative studies demonstrate that the accuracy of combined forecasts is significantly superior to those obtained from individual NN models.

Exploring the moderating role of growth options on the relation between board characteristics and management earnings forecasts

Manuscript Type
Empirical
Research Question/Issue
This study examines whether director independence, reputation, and financial expertise are related to management earnings forecast (MEF) activity. In particular, we examine whether such a relationship is moderated by firms’ growth options.
Research Findings/Insights
Using Australian archival data for 1,928 firm-years between 1999 and 2006, we find several board characteristics have a significant positive relationship with: (1) the likelihood of firms issuing MEFs; (2) their specificity; (3) their accuracy; and (4) a negative relationship with their bias. For (1), (2), and (3) we show that these relationships are accentuated for firms with high growth options.
Theoretical/Academic Implications
While the theory of voluntary disclosure suggests firms will disclose information that is favorable to them or their managers, well-governed firms issue informative MEFs that potentially reduce information asymmetries in capital markets. We extend the prior literature by showing that such a relation is enhanced in the presence of information asymmetry and moral hazard associated with growth options.
Practitioner/Policy Implications
Our results have strategic implications for nomination committees by showing that independent directors and those with strong reputations and financial expertise enhance the governance of high growth firms. We also inform the regulatory debate by showing that good corporate governance enhancing disclosure quality is context-specific – it is not a case of “one size fits all”.

Combined nonparametric prediction intervals for wind power generation

Prediction intervals (PIs) are a promising tool for quantification of uncertainties associated with point forecasts of wind power. However, construction of PIs using parametric methods is questionable, as forecast errors do not follow a standard distribution. This paper proposes a nonparametric method for construction of reliable PIs for neural network (NN) forecasts. A lower upper bound estimation (LUBE) method is adapted for construction of PIs for wind power generation. A new framework is proposed for synthesizing PIs generated using an ensemble of NN models in the LUBE method. This is done to guard against NN performance instability in generating reliable and informative PIs. A validation set is applied for short listing NNs based on the quality of PIs. Then, PIs constructed using filtered NNs are aggregated to obtain combined PIs. Performance of the proposed method is examined using data sets taken from two wind farms in Australia. Simulation results indicate that the quality of combined PIs is significantly superior to the quality of PIs constructed using NN models ranked and filtered by the validation set.

Neural network and interval type-2 fuzzy system for stock price forecasting

Stock price forecast has long been received special attention of investors and financial institutions. As stock prices are changeable over time and increasingly uncertain in modern financial markets, their forecasting becomes more important than ever before. A hybrid approach consisting of two components, a neural network and a fuzzy logic system, is proposed in this paper for stock price prediction. The first component of the hybrid, i.e. a feedforward neural network (FFNN), is used to select inputs that are highly relevant to the dependent variables. An interval type-2 fuzzy logic system (IT2 FLS) is employed as the second component of the hybrid forecasting method. The IT2 FLS’s parameters are initialized through deployment of the k-means clustering method and they are adjusted by the genetic algorithm. Experimental results demonstrate the efficiency of the FFNN input selection approach as it reduces the complexity and increase the accuracy of the forecasting models. In addition, IT2 FLS outperforms the widely used type-1 FLS and FFNN models in stock price forecasting. The combination of the FFNN and the IT2 FLS produces dominant forecasting accuracy compared to employing only the IT2 FLSs without the FFNN input selection.

Bayesian model averaging of load demand forecasts from neural network models

Creating a set of a number of neural network (NN) models in an ensemble and accumulating them can achieve better overview capability as compared to single neural network. Neural network ensembles are designed to provide solutions to particular problems. Many researchers and academicians have adopted this NN ensemble technique, especially in machine learning, and has been applied in various fields of engineering, medicine and information technology. This paper present a robust aggregation methodology for load demand forecasting based on Bayesian Model Averaging of a set of neural network models in an ensemble. This paper estimate a vector of coefficient for individual NN models' forecasts using validation data-set. These coefficients, also known as weights, are equal to posterior probabilities of the models generating the forecasts. These BMA weights are then used in combining forecasts generated from NN models with test data-set. By comparing the Bayesian results with the Simple Averaging method, it was observed that benefits are obtained by utilizing an advanced method like BMA for forecast combinations.

Variance-covariance based weighing for neural network ensembles

Neural network (NN) is a popular artificial intelligence technique for solving complicated problems due to their inherent capabilities. However generalization in NN can be harmed by a number of factors including parameter's initialization, inappropriate network topology and setting parameters of the training process itself. Forecast combinations of NN models have the potential for improved generalization and lower training time. A weighted averaging based on Variance-Covariance method that assigns greater weight to the forecasts producing lower error, instead of equal weights is practiced in this paper. While implementing the method, combination of forecasts is done with all candidate models in one experiment and with the best selected models in another experiment. It is observed during the empirical analysis that forecasting accuracy is improved by combining the best individual NN models. Another finding of this study is that reducing the number of NN models increases the diversity and, hence, accuracy.

Flood forecasting in large rivers with data-driven models

Results from the application of adaptive neuro-fuzzy inference system (ANFIS) to forecast water levels at 3 stations along the mainstream of the Lower Mekong River are reported in this paper. The study investigated the effects of including water levels from upstream stations and tributaries, and rainfall as inputs to ANFIS models developed for the 3 stations. When upstream water levels in the mainstream were used as input, improvements to forecasts were realized only when the water levels from 1 or at most 2 upstream stations were included. This is because when there are significant contributions of flow from the tributaries, the correlation between the water levels in the upstream stations and stations of interest decreases, limiting the effectiveness of including water levels from upstream stations as inputs. In addition, only improvements at short lead times were achieved. Including the water level from the tributaries did not significantly improve forecast results. This is attributed mainly to the fact that the flow contributions represented by the tributaries may not be significant enough, given that there could be large volume of flow discharging directly from the catchments which are ungauged, into the mainstream. The largest improvement for 1-day forecasts was obtained for Kratie station where lateral flow contribution was 17 %, the highest for the 3 stations considered. The inclusion of rainfall as input resulted in significant improvements to long-term forecasts. For Thakhek, where rainfall is most significant, the persistence index and coefficient of efficiency for 5-lead-day forecasts improved from 0.17 to 0.44 and 0.89 to 0.93, respectively, whereas the root mean square error decreased from 0.83 to 0.69 m.

Market effects on forecasting construction prices using vector error correction models

Construction price forecasting is an essential component to facilitate decision-making for construction contractors, investors and related financial institutions. Construction economists are increasingly interested in seeking a more analytical method to forecast construction prices. Although many studies have focused on construction price modelling and forecasting, few have considered the impacts of large-scale economic events and seasonality. In this study, an advanced multivariate modelling technique, namely the vector correction (VEC) model with dummy variables, was employed. The impacts of global economic events and seasonality are factored into the model to forecast the construction price in the Australian construction market. Research findings suggest that both long-run and dynamic short-term causal relationships exist among the price and levels of supply and demand in the construction market. These relationships drive the construction price and supply and demand, which interact with one another as a loop system. The reliability of forecasting models was examined by the mean absolute percentage error (MAPE) and the Theil's inequality coefficient U tests. The test results suggest that the conventional VEC model and the VEC model with dummy variable are both acceptable for forecasting the construction price, while the VEC model considering external impacts achieves higher prediction accuracy than the conventional VEC model. © 2014 © 2014 Taylor & Francis.

A panel vector error correction approach to forecasting demand in regional construction markets

Reliable forecasting as to the level of aggregate demand for construction is of vital importance to developers, builders and policymakers. Previous construction demand forecasting studies mainly focused on temporal estimating using national aggregate data. The construction market can be better represented by a group of interconnected regions or local markets rather than a national aggregate, and yet regional forecasting techniques have rarely been applied. Furthermore, limited research has applied regional variations in construction markets to construction demand modelling and forecasting. A new comprehensive method is used, a panel vector error correction approach, to forecast regional construction demand using Australia’s state-level data. The links between regional construction demand and general economic indicators are investigated by panel cointegration and causality analysis. The empirical results suggest that both long-run and causal links are found between regional construction demand and construction price, state income, population, unemployment rates and interest rates. The panel vector error correction model can provide reliable and robust forecasting with less than 10% of the mean absolute percentage error for a medium-term trend of regional construction demand and outperforms the conventional forecasting models (panel multiple regression and time series multiple regression model). The key macroeconomic factors of construction demand variations across regions in Australia are also presented. The findings and robust econometric techniques used are valuable to construction economists in examining future construction markets at a regional level.

Least squares learning and the US Treasury bill rate

Understanding how agents formulate their expectations about Fed behavior is important for market participants because they can potentially use this information to make more accurate estimates of stock and bond prices. Although it is commonly assumed that agents learn over time, there is scant empirical evidence in support of this assumption. Thus, in this paper we test if the forecast of the three month T-bill rate in the Survey of Professional Forecasters (SPF) is consistent with least squares learning when there are discrete shifts in monetary policy. We first derive the mean, variance and autocovariances of the forecast errors from a recursive least squares learning algorithm when there are breaks in the structure of the model. We then apply the Bai and Perron (1998) test for structural change to a forecasting model for the three month T-bill rate in order to identify changes in monetary policy. Having identified the policy regimes, we then estimate the implied biases in the interest rate forecasts within each regime. We find that when the forecast errors from the SPF are corrected for the biases due to shifts in policy, the forecasts are consistent with least squares learning. © 2014 Elsevier B.V.

Is carbon emissions trading profitable?

The European Carbon Emissions Trading Scheme introduced in 2005 has led to both spot and futures market trading of carbon emissions. However, despite seven years of trading, we have no knowledge on how profitable carbon emissions trading is. In this paper, we first test whether carbon forward returns predict carbon spot returns. We find strong evidence on both in-sample and out-of-sample predictability. Based on this evidence, we forecast carbon spot returns using both carbon forward returns and a constant. We consider a mean-variance investor and a CRRA investor, and show that they have higher utility and can make more statistically significant profits by following forecasts generated from the forward returns model than from a constant returns model.

Corrigendum to “Least squares learning and the US treasury bill rate” [Econ. Syst. 38 (2014) 194–204]

Understanding how agents formulate their expectations about Fed behavior is important for market participants because they can potentially use this information to make more accurate estimates of stock and bond prices. Although it is commonly assumed that agents learn over time, there is scant empirical evidence in support of this assumption. Thus, in this paper we test if the forecast of the three month T-bill rate in the Survey of Professional Forecasters (SPF) is consistent with least squares learning when there are discrete shifts in monetary policy. We first derive the mean, variance and autocovariances of the forecast errors from a recursive least squares learning algorithm when there are breaks in the structure of the model. We then apply the Bai and Perron (1998) test for structural change to a forecasting model for the three month T-bill rate in order to identify changes in monetary policy. Having identified the policy regimes, we then estimate the implied biases in the interest rate forecasts within each regime. We find that when the forecast errors from the SPF are corrected for the biases due to shifts in policy, the forecasts are consistent with least squares learning.