Can we fruitfully use the same macroeconomic model to forecast and to perform policy analysis? There is a tension between a model’s ability to forecast accurately and its ability to tell a theoretically consistent story. The aim of this dissertation is to propose ways to soothe this tension, combining structural and reduced-form models in order to have models that can effectively do both.

Syftet med föreliggande studie var att försöka reda i litteraturen kring två välkända begrepp inom Affective Forecasting nämligen Impact Bias, som innebär att människor har en tendens att överskatta i vilken utsträckning de kommer att uppleva en viss känsla i en framtida situation än vad som senare visar sig vara fallet, och Empathy Gaps, som innebär att människor har en tendens att underskatta i vilken grad känslotillstånd kommer att påverka deras preferenser i en framtida situation samt pröva dessa begrepp i en och samma enkätundersökning. Etthundra sextiotvå studenter, slumpvist uppdelade i två grupper, Känslogrupp och Preferensgrupp, deltog frivilligt i undersökningen. Enkätundersökningen var en mixad design med grupp (känsla kontra preferens) som mellangruppsfaktor och förtest kontra eftertest som inomgruppsfaktor. I studien visade samtliga gruppers resultat i linje med Impact Bias teorin, dvs. att deltagarna i både Känslogrupp och Preferensgrupp skattade lägre i eftertest (actual) än pretest (forecasting). Resultatet diskuteras bla. utifrån Construal Level Theory, CLT. Förslag på vidare forskning ges.

This thesis is focused on investigating the predictability of exchange rate returns on monthly and daily frequency using models that have been mostly developed in the machine learning field. The forecasting performance of these models will be compared to the Random Walk, which is the benchmark model for financial returns, and the popular autoregressive process. The machine learning models that will be used are Regression trees, Random Forests, Support Vector Regression (SVR), Least Absolute Shrinkage and Selection Operator (LASSO) and Bayesian Additive Regression trees (BART). A characterizing feature of financial returns data is the presence of volatility clustering, i.e. the tendency of persistent periods of low or high variance in the time series. This is in disagreement with the machine learning models which implicitly assume a constant variance. We therefore extend these models with the most widely used model for volatility clustering, the Generalized Autoregressive Conditional Heteroscedasticity (GARCH) process. This allows us to jointly estimate the time varying variance and the parameters of the machine learning using an iterative procedure. These GARCH-extended machine learning models are then applied to make one-step-ahead prediction by recursive estimation that the parameters estimated by this model are also updated with the new information. In order to predict returns, information related to the economic variables and the lagged variable will be used. This study is repeated on three different exchange rate returns: EUR/SEK, EUR/USD and USD/SEK in order to obtain robust results. Our result shows that machine learning models are capable of forecasting exchange returns both on daily and monthly frequency. The results were mixed, however. Overall, it was GARCH-extended SVR that shows great potential for improving the predictive performance of the forecasting of exchange rate returns.

In recent years the electric power industry has been challenged by a high level of uncertainty and volatility brought on by deregulation and globalization. A power producer must minimize the life cycle cost while meeting stringent safety and regulatory requirements and fulfilling customer demand for high reliability. Therefore, to achieve true system excellence, a more sophisticated system-level decision-making process with a more accurate forecasting support system to manage diverse and often widely dispersed generation units as a single, easily scaled and deployed fleet system in order to fully utilize the critical assets of a power producer has been created as a response. The process takes into account the time horizon for each of the major decision actions taken in a power plant and develops methods for information sharing between them. These decisions are highly interrelated and no optimal operation can be achieved without sharing information in the overall process. The process includes a forecasting system to provide information for planning for uncertainty. A new forecasting method is proposed, which utilizes a synergy of several modeling techniques properly combined at different time-scales of the forecasting objects. It can not only take advantages of the abundant historical data but also take into account the impact of pertinent driving forces from the external business environment to achieve more accurate forecasting results. Then block bootstrap is utilized to measure the bias in the estimate of the expected life cycle cost which will actually be needed to drive the business for a power plant in the long run. Finally, scenario analysis is used to provide a composite picture of future developments for decision making or strategic planning. The decision-making process is applied to a typical power producer chosen to represent challenging customer demand during high-demand periods. The process enhances system excellence by providing more accurate market information, evaluating the impact of external business environment, and considering cross-scale interactions between decision actions. Along with this process, system operation strategies, maintenance schedules, and capacity expansion plans that guide the operation of the power plant are optimally identified, and the total life cycle costs are estimated.

This study attempts to capture the behavior of volatility in the commodity futures market by importing the normal mixture GARCH Model and the NIG mixture GARCH model (Normal-inverse Gaussian Mixture GARCH Model). Normal mixture GARCH Model (what follows called NM-GARCH Model) is a model mixed by two to several normal distributions with a specific weight portfolio, and its variance abide by GAECH process. The ability of capturing the financial data with leptokurtosis and fat-tail of NM-GARCH Model is better than Normal GARCH Model and Student¡¦s t GARCH Model.¡CAlso¡AThe Variance of the factor with lower weight in NM-GARCH Model usually higher, and the volatility of the factor with higher weight is lower, which explains the situation happens in the real market that the probability of large fluctuations (shocks) is small, and the probability of small fluctuations are higher. Generally, the volatilities which keeping occurring in common cases are respectively flat, and the shocks usually bring large impacts but less frequent. NIG Mixture Distribution is a distribution mixed by two to several weighted distributions, and the distribution of every factor abides by NIG Distribution. Compare to Normal Mixture Distribution, NIG Mixture Distribution takes the advantages of NIG Distribution into account, which can not only explain leptokurtosis and the deviation of data, but describe the fat-tail phenomenon more complete as well, because of the both tails of NIG Distribution decreasing slowly. This study will apply the NM GARCH Model and NIG GARCH Model to the Volatility forecasting of the return rates in the crude oil futures market, and infer the predictive abilities of this two kinds of models are significantly better than other volatility model by implementing parameter estimation, forecasting, loss function and statistic significant test.

No description available.

The objective of this thesis is to study the load forecasting of distribution feeders and substations for Fong-Shan District of Taiwan Power Company. To increase the accuracy of load forecasting, the load characterization of customers served has been investigated. The typical load patterns of different customers classes and derived by performing the statistic of power consumption data retrieved. The daily load profiles and load consumptions data distribution feeders and substations have been solved by considering the typical load patterns and energy consumption of all customers served. To investigate the correlation ship of temperature and energy consumption of customer classes, the temperature sensitivity of customer energy consumption has been used to update the load composition and the contribution of load change by different customer classes. To perform the load forecasting of distribution systems, the linear, nonlinear and hybrid load forecasting modules have been proposed. The historical load data of distribution feeders and substations in Fong-Shan District have been used to derive the load forecasting modules. To analyze the accuracy of load forecasting by considering the temperature effect, the temperature change is included in the load forecasting module. With the load forecasting derived, the proper load transfers among different distribution feeders and different substations have been determined to achieve the load balancing of service areas.

none

This study examines the dynamics and predictability of the mesoscale convective vortex (MCV) event of 10-13 June 2003 which occurred during the Bow Echo and Mesoscale Convective Vortex Experiment (BAMEX). The MCV formed from a preexisting upper-level disturbance over the southwest United States on 10 June and matured as it traveled northeastward. The BAMEX field campaign provided a relatively dense collection of upper air observations through dropsondes on 11 June during the mature stage of the vortex. While several previous studies have focused on analysis of the dynamics and thermodynamics of observed and simulated vortices, few have addressed the ability to predict MCVs using numerical models. This event is of particular interest to the study of MCV dynamics and predictability given the anomalously strong and long-lived nature of the circulation and the dense data set. The first part of this study explores the dynamics of this MCV through an in-depth analysis of data from the profiler network and BAMEX dropsonde observations, in addition to the conventional surface and sounding observations as well as radar and satellite images. Next, issues relating to model performance are addressed through anevaluation of two state-of-the-art mesoscale models with varying resolutions. It is determined that the ability of a forecast model to accurately predict this MCV event is directly related to its ability to simulate convection. It is also shown that the convective-resolving Weather Research and Forecast (WRF) model with horizontal grid increments of 4 km displays superior performance in its simulation of this MCV event. Finally, an ensemble of 20 forecasts using mesoscale model MM5 with horizontal grid increments of 10 km are employed to evaluate probabilistically the dynamics and predictability of the MCV through the examination of the ensemble spread as well as the correlations between different forecast variables among ensemble members. It is shown that after MCV development, the ensemble mean performs poorly while individual ensemble members with good forecasts of convection at all stages of the MCV also forecast the midlevel vortex well. Furthermore, correlations among ensemble members generally support the findings in the observational analysis and in previous literature.

This dissertation consists of three essays. Chapter II uses the method of structural factor analysis to study the effects of monetary policy on key macroeconomic variables in a data rich environment. I propose two structural factor models. One is the structural factor augmented vector autoregressive (SFAVAR) model and the other is the structural factor vector autoregressive (SFVAR) model. Compared to the traditional vector autogression (VAR) model, both models incorporate far more information from hundreds of data series, series that can be and are monitored by the Central Bank. Moreover, the factors used are structurally meaningful, a feature that adds to the understanding of the “black box” of the monetary transmission mechanism. Both models generate qualitatively reasonable impulse response functions. Using the SFVAR model, both the “price puzzle” and the “liquidity puzzle” are eliminated. Chapter III employs the method of structural factor analysis to conduct a forecasting exercise in a data rich environment. I simulate out-of-sample real time forecasting using a structural dynamic factor forecasting model and its variations. I use several structural factors to summarize the information from a large set of candidate explanatory variables. Compared to Stock and Watson (2002)’s models, the models proposed in this chapter can further allow me to select the factors structurally for each variable to be forecasted. I find advantages to using the structural dynamic factor forecasting models compared to alternatives that include univariate autoregression (AR) model, the VAR model and Stock and Watson’s (2002) models, especially when forecasting real variables. In chapter IV, we measure U.S. technology shocks by implementing a dual approach, which is based on more reliable price data instead of aggregate quantity data. By doing so, we find the relative volatility of technology shocks and the correlation between output fluctuation and technology shocks to be much smaller than those revealed in most real-business-cycle (RBC) studies. Our results support the findings of Burnside, Eichenbaum and Rebelo (1996), who showed that the correlation between technology shocks and output is exaggerated in the RBC literature. This suggests that one should examine other sources of fluctuations for a better understanding of the business cycle phenomena.