There are many models used to estimate demand elasticities. We present a complex review of these studies in our thesis. Our empirical goal is to compare LES, Translog and QUAIDS demand systems according to their performance. In parallel, we estimate the elasticities of meat demand in the Czech Republic for the period 2010 - 2015 using the data of the household budget survey. Comparing the systems by the Akaike and Schwarz criterion, LES demonstrates the best fit for this kind of data. The average of price elasticity for different kinds of meat in the examined period is -0.99, income elasticity then equals to 1.12. These results can have important implications for tax policy, or for commercial use. JEL Classification F12, F21, F23, H25, H71, H87 Keywords Demand, comparison, LES, Translog, QUAIDS, meat Author's e-mail 55606678@fsv.cuni.cz Supervisor's e-mail milan.scasny@czp.cuni.cz

This thesis describes a new method to evaluate the value of wind power combined with domestic demand response. The thesis gives a brief overview of current domestic demand management programmes, and highlights the demand response and its current application. Such technology has conventionally been used for different purposes, such as frequency regulation, and to minimize the spot electricity prices in the market. The aim is to show whether such technology may become useful to make the renewables, and in particular wind power more interesting for investors. An assessment framework based on generation scheduling is developed to quantify the value of wind power. A further important aspect of value of wind power is the impact of intermittency on overall reliability of the system. This necessitates increasing the spinning reserve level which will increase the production cost. The changes in the spinning reserve level has been investigated in this thesis and it has been shown that how different forecasting errors may change the overall value of a windfarm over its lifetime. One of the most important aspects of a system containing demand response, is the availability of demand response. A load modelling package is developed to show the potential for demand response in a real system from domestic sector. With every increasing the concerns with regard to future of generation mix in Britain, this work has proposed over 72 scenarios for the future of generation mix in Britain and the impact of demand response to increase the value of wind power in 2020 has been investigated. The assessment framework is enhanced by showing that how the value of wind power combined with domestic demand response may change by changes in emission price, and cost of demand response. This will show the degree of feasibility of such system in which demand response is treated like a commodity.

During the period 1996 to 2003 the number of GP services per capita in Australia fell by 14 percent and the proportion of services bulk billed (ie provided at no cost to the patient) fell by 12 percentage points. The Government responded to these trends by outlaying hundreds of millions of dollars to increase Government medical insurance rebates, to increase the number of GPs in Australia, and to provide incentives for GPs to bulk bill. ¶ There has been no comprehensive modelling of the GP market to assist in understanding the reasons for either the declining trends or whether the Government responses were successful. This thesis aims to fill that gap. ¶ Previous Australian modelling of the GP market has been cross sectional and mostly demand focused. This thesis uses panel data to minimise the biases caused by unobserved heterogeneity and border crossing, and to estimate explicit supply and demand equations to enable the relationship between supply and demand to be explored. ¶ This approach estimates the impact on GP market outcomes of both policy decisions regarding rebates and GP numbers, and of external changes such as the trends in social attitudes and age. The likely future paths of the market without further policy change can be considered, and the measures needed to meet given policy targets determined. ¶ In addressing these questions it is also shown that supplier induced demand does apply in Australian general practice but is not material, that previous cross sectional analysis was biased due to border crossing by patients, that GPs who charge patients with concession cards less than other patients are behaving economically rationally, and that when the Government increases the Medicare rebate payment, 85% of the increase goes to the GP and 15% to the patient. The analysis also shows that GP density has no significant effect on mortality in Australia, and was unable to detect any effect of the business cycle on mortality. ¶ The demand curve for Australian general practice services is shown to be fundamentally determined by the real value of the MBS rebate in the short term, where the real value adjusted for growth in average weekly earnings. ¶ The supply curve is determined by aggregate numbers of GPs and by the number of services they each provide. The average number of services provided per GP is determined by GP age and gender, but more importantly by a trend effect thought to be due to attitudinal changes which must be explored further, and must be incorporated into any prediction of GP market outcomes. ¶ The thesis provides the first empirically based overview of the behaviour of the GP market at end of the twentieth century, and shows how Government policy levers and other trends interact to generate the market outcomes. If the Government has targets for service levels or charging patterns in general practice, these models can facilitate determination of the policy options appropriate to achieve those targets.

Energy consumption in the residential sector of Ontario is expected to grow by 15%, most of which is expected to be from electricity use, with an annual average growth rate of 0.9% between 2010 and 2020. With Ontario government’s Integrated Power System Plan (IPSP) recommending phasing out coal fired generators by 2014, the execution of Conservation and Demand Management and Demand Response programs can have significant impact on reducing power consumption and peak demand in the province. Electricity generation, especially from fossil fuel, contributes 18% of total green house gas (GHG) emissions in Ontario. With climate change effects being attributed to GHG emissions and environmental regulations, it is necessary to reduce GHG emissions from power generation sector. In this context, the current Energy Hub Management System project, of which the work presented here is a part, may lead to the reduction of electricity power demand and GHG emissions in Ontario. This thesis presents the validation of Energy Hub Management System (EHMS) residential sector model. Performances of individual appliances and the results obtained from various case-studies considering the EHMS model are compared with respect to a base case representing a typical residential customer. The case-studies are carefully developed to demonstrate the capability of the EHMS model to generate optimum operational schedules to minimize energy costs, energy consumption and emissions based on user defined constraints and preferences. Furthermore, a forecasting methodology based on single variable econometric time series is developed to estimate day-ahead CO2 emissions from Ontario’s power generation sector. The forecasted emissions profile is integrated into the EHMS model to optimize a residential customer’s contribution to CO2 emissions in Ontario.

Energy consumption in the residential sector of Ontario is expected to grow by 15%, most of which is expected to be from electricity use, with an annual average growth rate of 0.9% between 2010 and 2020. With Ontario government’s Integrated Power System Plan (IPSP) recommending phasing out coal fired generators by 2014, the execution of Conservation and Demand Management and Demand Response programs can have significant impact on reducing power consumption and peak demand in the province. Electricity generation, especially from fossil fuel, contributes 18% of total green house gas (GHG) emissions in Ontario. With climate change effects being attributed to GHG emissions and environmental regulations, it is necessary to reduce GHG emissions from power generation sector. In this context, the current Energy Hub Management System project, of which the work presented here is a part, may lead to the reduction of electricity power demand and GHG emissions in Ontario. This thesis presents the validation of Energy Hub Management System (EHMS) residential sector model. Performances of individual appliances and the results obtained from various case-studies considering the EHMS model are compared with respect to a base case representing a typical residential customer. The case-studies are carefully developed to demonstrate the capability of the EHMS model to generate optimum operational schedules to minimize energy costs, energy consumption and emissions based on user defined constraints and preferences. Furthermore, a forecasting methodology based on single variable econometric time series is developed to estimate day-ahead CO2 emissions from Ontario’s power generation sector. The forecasted emissions profile is integrated into the EHMS model to optimize a residential customer’s contribution to CO2 emissions in Ontario.

Two broad surveys of community- level water consumption and pricing behavior are used to answer questions about water demand in a more flexible and dynamic context than is provided in the literature. Central themes of price representation, aggregation, and dynamic adjustment tie together three econometric demand analyses. The centerpiece of each analysis is an exogenous weighted price representation. A model in first-differences is estimated by ordinary least squares using data from a personally-conducted survey of Texas urban water suppliers. Annual price elasticity is found to vary with weather and income, with a value of -0.127 at the data mean. The dynamic model becomes a periodic error correction model when the residuals of 12 static monthly models are inserted into the difference model. Distinct residential, commercial, and industrial variables and historical climatic conditions are added to the integrated model, using new national data. Quantity demanded is found to be periodically integrated with a common stochastic root. Because of this, the structural monthly models must be cointegrated to be consistent, which they appear to be. The error correction coefficient is estimated at -0.187. Demand is found to be seasonal and slow to adjust to shocks, with little or no adjustment in a single year and 90% adjustment taking a decade or more. Residential and commercial demand parameters are found to be indistinguishable. The sources of price endogeneity and historical fixes are reviewed. Ideal properties of a weighted price index are identified. For schedules containing exactly two rates, weighting is equivalent to a distribution function in consumption. This property is exploited to derive empirical weights from the national data, using values from a nonparametric generalization of the structural demand model and a nonparametric cumulative density function. The result is a generalization of the price difference metric to a weighted level-price index. The validity of a uniform weighting is not rejected. The weighted price index is data intensive, but the payoff is increased depth and precision for the economist and accessibility for the practitioner.

Discrete choice models (DCMs) are used to forecast demand in a variety of engineering, marketing, and policy contexts, and understanding the uncertainty associated with model forecasts is crucial to inform decision-making. This thesis evaluates the suitability of DCMs for forecasting automotive demand. The entire scope of this investigation is too broad to be covered here, but I explore several elements with a focus on three themes: defining how to measure forecast accuracy, comparing model specifications and forecasting methods in terms of prediction accuracy, and comparing the implications of model specifications and forecasting methods on vehicle design. Specifically I address several questions regarding the accuracy and uncertainty of market share predictions resulting from choice of utility function and structural specification, estimation method, and data structure assumptions. I1 compare more than 9,000 models based on those used in peer-reviewed literature and academic and government studies. Firstly, I find that including more model covariates generally improves predictive accuracy, but that the form those covariates take in the utility function is less important. Secondly, better model fit correlates well with better predictive accuracy; however, the models I construct— representative of those in extant literature— exhibit substantial prediction error stemming largely from limited model fit due to unobserved attributes. Lastly, accuracy of predictions in existing markets is neither a necessary nor sufficient condition for use in design. Much of the econometrics literature on vehicle market modeling has presumed that biased coefficients make for bad models. For purely predictive purposes, the drawbacks of potentially mitigating bias using generalized method of moments estimation coupled with instrumental variables outweigh the expected benefits in the experiments conducted in this dissertation. The risk of specifying invalid instruments is high, and my results suggest that the instruments frequently used in the automotive demand literature are likely invalid. Furthermore, biased coefficients are not necessarily bad for maximizing the predictive power of the model. Bias can even aid predictions by implicitly capturing persistent unobserved effects in some circumstances. Including alternative specific constants (ASCs) in DCM utility functions improves model fit but not necessarily forecast accuracy. For frequentist estimated models all tested methods of forecasting ASCs improved share predictions of the whole midsize sedan market over excluding ASC in predictions, but only one method results in improved long term new vehicle, or entrant, forecasts. As seen in a synthetic data study, assuming an incorrect relationship between observed attributes and the ASC for forecasting risks making worse forecasts than would be made by a model that excludes ASCs entirely. Treating the ASCs as model parameters with full distributions of uncertainty via Bayesian estimation is more robust to selection of ASC forecasting method and less reliant on persistent market structures, however it comes at increased computational cost. Additionally, the best long term forecasts are made by the frequentist model that treats ASCs as calibration constants fit to the model post estimation of other parameters.

This thesis describes a short term water demand forecasting application that is based upon a combination of a neural network forecast generator and a rule based system that modifies the resulting forecasts. Conventionally, short term forecasting of both water consumption and electrical load demand has been based upon mathematical models that aim to either extract the mathematical properties displayed by a time series of historical data, or represent the causal relationships between the level of demand and the key factors that determine that demand. These conventional approaches have been able to achieve acceptable levels of prediction accuracy for those days where distorting, non cyclic influences are not present to a significant degree. However, when such distortions are present, then the resultant decrease in prediction accuracy has a detrimental effect upon the controlling systems that are attempting to optimise the operation of the water or electricity supply network. The abnormal, non cyclic factors can be divided into those which are related to changes in the supply network itself, those that are related to particular dates or times of the year and those which are related to the prevailing meteorological conditions. If a prediction system is to provide consistently accurate forecasts then it has to be able to incorporate the effects of each of the factor types outlined above. The prediction system proposed in this thesis achieves this by the use of a neural network that by the application of appropriately classified example sets, can track the varying relationship between the level of demand and key meteorological variables. The influence of supply network changes and calendar related events are accounted for by the use of a rule base of prediction adjusting rules that are built up with reference to past occurrences of similar events. The resulting system is capable of eliminating a significant proportion of the large prediction errors that can lead to non optimal supply network operation.

Thesis (Ph.D)--Industrial and Systems Engineering, Georgia Institute of Technology, 2009. / Committee Chair: Keskinocak, Pinar; Committee Co-Chair: Vande Vate, John; Committee Member: Ferguson, Mark; Committee Member: Griffin, Paul; Committee Member: Swann, Julie. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Thesis (Ph. D.)--New York University, 1989. / Includes bibliographical references (leaves 165-182).