The main purpose of the study was to produce a population forecast by sex and one-year age groups for the Republic of Azerbaijan for the period 2009-2050. The forecasting process consisted from the several stages, which are included in the general structure of the production process of a forecast. During the data analysis, a number of methods were applied: redistributional methods of intensities, method of reconstruction abridge life tables into complete life tables, application of the Kannisto method to describe mortality in the old ages. The estimated one-year intensities were utilized in the current trends analysis serving as a background for forecasting the parameters of the cohort-component projection model. During the population forecasting the cohort-component projection model was used and population was forecasted in three scenarios (variants). Underlying assumptions were based on analysis of the recent demographic trends and an assessment of their implications for future changes. Keywords: forecast, population development, Azerbaijan

The accuracy and quality of river forecasts are dependent on the nature of each flood. Less extreme , more common, floods may afford deviations between the predicted forecast and observed stage because the locals may be prepared, based on past experience to deal with the less extreme flood events. For less frequent, high flow events the flood forecasts and advanced warning time are more critical, because the locals need time to develop emergency response plans. The National Weather Service River Forecast Centers (NWS RFC) develop the river forecasts and provide them to the National Weather Service Weather Forecast Office (NWS WFO) for dissemination. During flood events the RFC's are tasked with processing the observed data and running, reviewing and modifying the forecast models to provide reasonable river forecasts based on observed conditions and the forecasters' experience. This thesis will discuss the personal experiences of the author, analyze the components of the National Weather Service river forecasting process, analyze June 2008 river and precipitation forecasts for several eastern Iowa watersheds, and discuss the results of the analysis as well as provide support to current calls to action to support forecast verification through the hindcasting process.

<p>In this paper we aim to investigate volatility and correlation of sector</p><p>indexes of Nordic Market. More precisely we work with OMX Stockholm</p><p>Exchange Indexes, considering the Paper, the Energy and the Bank</p><p>sectors.</p><p>We use daily returns over the period from 5 January 2001 to 13 April</p><p>2007 and compute and forecast return volatility using the GARCH(1; 1)</p><p>model. We also calculate the correlation matrix of the indexes.</p><p>The GARCH(1; 1) model ¯t the empirical data well for all three sectors</p><p>and can therefore be used for volatility forecasts. Here, we have pre-</p><p>dicted the one-day-ahead forecasts and based on these data calculated</p><p>the correlation matrix. The results from these calculations show that</p><p>all three sectors are highly correlated. We obtained however the small-</p><p>est correlation between Paper and Energy which was surprising as the</p><p>Paper industry is very energy consuming. This result indicates other</p><p>relations between Paper and Energy.</p>

In this paper we aim to investigate volatility and correlation of sector indexes of Nordic Market. More precisely we work with OMX Stockholm Exchange Indexes, considering the Paper, the Energy and the Bank sectors. We use daily returns over the period from 5 January 2001 to 13 April 2007 and compute and forecast return volatility using the GARCH(1; 1) model. We also calculate the correlation matrix of the indexes. The GARCH(1; 1) model ¯t the empirical data well for all three sectors and can therefore be used for volatility forecasts. Here, we have pre- dicted the one-day-ahead forecasts and based on these data calculated the correlation matrix. The results from these calculations show that all three sectors are highly correlated. We obtained however the small- est correlation between Paper and Energy which was surprising as the Paper industry is very energy consuming. This result indicates other relations between Paper and Energy.

This thesis consists of three essays in empirical finance and macroeconomics. The first essay proposes a new structural-break vector autoregressive model for predicting real output growth by the nominal yield curve. The model allows for the possibility of both in-sample and out-of-sample breaks in parameter values and uses information in historical regimes to make inference on out-of-sample breaks. A Bayesian estimation and forecasting procedure is developed which accounts for the uncertainty of both structural breaks and model parameters. I discuss dynamic consistency when forecasting recursively and provide a solution. Applied to monthly US data, I find strong evidence of breaks in the predictive relation between the yield curve and output growth. Incorporating the possibility of structural breaks improves out-of-sample forecasts of output growth. The second essay proposes a sequential Monte Carlo method for estimating GARCH models subject to an unknown number of structural breaks. We use particle filtering techniques that allow for fast and efficient updates of posterior quantities and forecasts in real-time. The method conveniently deals with the path dependence problem that arises in these type of models. The performance of the method is shown to work well using simulated data. Applied to daily NASDAQ returns, we find strong evidence of structural breaks in the long-run variance of returns. Models with flexible return distributions such as t-innovations or with jumps indicate fewer breaks than models with normal return innovations and are favored by the data. The third essay proposes a new tilt stochastic volatility model which extends the existing volatility models by modeling the asymmetric correlation between return and volatility innovations in a unified and flexible framework. The Efficient Importance Sampling (EIS) procedure is adapted to estimate the model. Simulation studies show that the Maximum Likelihood (ML)-EIS estimation of the model is accurate. The new model is applied to the CRSP daily returns. I find the extensions are significant and incorporating them improves the accuracy of volatility estimates.

This thesis consists of three essays in empirical finance and macroeconomics. The first essay proposes a new structural-break vector autoregressive model for predicting real output growth by the nominal yield curve. The model allows for the possibility of both in-sample and out-of-sample breaks in parameter values and uses information in historical regimes to make inference on out-of-sample breaks. A Bayesian estimation and forecasting procedure is developed which accounts for the uncertainty of both structural breaks and model parameters. I discuss dynamic consistency when forecasting recursively and provide a solution. Applied to monthly US data, I find strong evidence of breaks in the predictive relation between the yield curve and output growth. Incorporating the possibility of structural breaks improves out-of-sample forecasts of output growth. The second essay proposes a sequential Monte Carlo method for estimating GARCH models subject to an unknown number of structural breaks. We use particle filtering techniques that allow for fast and efficient updates of posterior quantities and forecasts in real-time. The method conveniently deals with the path dependence problem that arises in these type of models. The performance of the method is shown to work well using simulated data. Applied to daily NASDAQ returns, we find strong evidence of structural breaks in the long-run variance of returns. Models with flexible return distributions such as t-innovations or with jumps indicate fewer breaks than models with normal return innovations and are favored by the data. The third essay proposes a new tilt stochastic volatility model which extends the existing volatility models by modeling the asymmetric correlation between return and volatility innovations in a unified and flexible framework. The Efficient Importance Sampling (EIS) procedure is adapted to estimate the model. Simulation studies show that the Maximum Likelihood (ML)-EIS estimation of the model is accurate. The new model is applied to the CRSP daily returns. I find the extensions are significant and incorporating them improves the accuracy of volatility estimates.

In a more competitive environment, load forecast serves two different applications. First, load forecast results can be used by the retailers of power to study their opportunities and plan their business strategies. Second, accurate projections of load are useful for T&D operators in performing system operation and expansion studies. Several key elements in their market and system planning studies have strong location factors that the spatial load forecast can address. In this dissertation, a package that integrates a Geographic Information System (GIS) used for automatic mapping and facility management (AM/FM) and a spatial load forecast module is presented. The interface functions and the procedure of the fuzzy logic based spatial load forecast module are described. Simulation studies are performed on a metropolitan area of Kaohsiung, Taiwan. The conventional fuzzy modeling has a drawback in that the fuzzy rules or the fuzzy membership functions are determined by trial and error. In this dissertation an automatic model identification procedure is proposed to construct the fuzzy model for short-term load forecast. In this method an analysis of variance is used to identify the influential variables on the system load. To setup the fuzzy rules, a cluster estimation method is adopted to determine the number of rules and the membership functions of variables involved in the premises of the rules. A recursive least square method is then used to determine the coefficients in the conclusion parts of the rules. None of these steps involves nonlinear optimization and all steps have well-bounded computation time.

The Weather Channel (TWC) and AccuWeather (AWX) are leading providers of weather information to the general public. The purpose of this Master’s Report is to examine the wind speed forecasts made by these two providers and determine their reliability and accuracy. The data used within this report was collected over a 12-month period at 51 locations across the state of Texas. The locations were grouped according to wind power class, which ranged from Class 1 to Class 4. The length of the forecast period was 9 days for TWC and 14 days for AWX. It was found that the values forecasted by TWC were generally not well calibrated, but were never far from being perfectly calibrated and always demonstrated positive skill. The sharpness of TWC’s forecasts decreased consistently with lead time, allowing them to maintain a skill score greater than the climatological average throughout the forecast period. TWC tended to over-forecast wind speed in short term forecasts, especially within the lower wind power class regions. AWX forecasts were found to have positive skill the first 6 days of the forecasting period before becoming near zero or negative. AWX’s forecasts maintained a fairly high sharpness throughout the forecast period, which helped contribute to increasingly un-calibrated forecast values and negative skill in longer term forecasts. The findings within this report should help provide a better understanding of the wind forecasts made by TWC and AWX, and determine the strengths and weaknesses of both companies. / text

none

A new method is presented to provide automatic sequencing of multiple hydrodynamic models and automated analysis of model forecast uncertainty. A Hydrodynamic and oil spill model Python (HyosPy) wrapper was developed to run the hydrodynamic model, link with the oil spill, and visualize results. The HyosPy wrapper completes the following steps automatically: (1) downloads wind and tide data (nowcast, forecast and historical); (2) converts data to hydrodynamic model input; (3) initializes a sequence of hydrodynamic models starting at pre-defined intervals on a multi-processor workstation. Each model starts from the latest observed data, so that the multiple models provide a range of forecast hydrodynamics with different initial and boundary conditions reflecting different forecast horizons. As a simple testbed for integration strategies and visualization on Google Earth, a Runge-Kutta 4th order (RK4) particle transport tracer routine is developed for oil spill transport. The model forecast uncertainty is estimated by the difference between forecasts in the sequenced model runs and quantified by using statistics measurements. The HyosPy integrated system with wind and tide force is demonstrated by introducing an imaginary oil spill in Corpus Christi Bay. The results show that challenges in operational oil spill modeling can be met by leveraging existing models and web-visualization methods to provide tools for emergency managers. / text