<p>The purpose of this thesis is to identify a general model to forecast GDP growth for the Baltic States, Estonia, Latvia and Lithuania. If the model provides reliable results for these states, then the model should be able to forecast GDP growth for other countries of interest. Forecasts are made by using a reduced vector autoregressive (VAR) model. The VAR models make use of past values of Gross Domestic Product-Inflation-Unemployment as explanatory variables.</p><p>The performed forecasts have provided good results for horizons up to t+8. The forecasts for 2009 (t+12) are in line with those of several other actors. It is reasonable to assume that some of the forecasts for t+16 have reliable results. The Lithuanian forecast show a fall in GDP with 12.51 per cent in 2009 and a GDP growth of 4.23 per cent in 2010. The forecast for Estonia show that the GDP will decrease with 1.49 per cent in 2009 and 12.72 per cent in 2010. Finally the forecast for Latvia show a fall in GDP of 3.1 per cent in 2009 and 18 per cent in 2010. From the findings it is possible to conclude that the model provided reliable estimates of future levels of GDP for the Baltic States and the benchmark countries. This indicates that the model should be applicable on other countries of interest.</p>

We analyze the interaction between monetary policy in the US and the global economy proposing a new class of Bayesian global vector autoregressive models that accounts for time-varying parameters and stochastic volatility (TVP-SV-GVAR). Our results suggest that US monetary policy responds to shocks to the global economy, in particular to global aggregate demand and monetary policy shocks. On the other hand, US-based contractionary monetary policy shocks lead to persistent international output contractions and a drop in global inflation rates, coupled with rising interest rates in advanced economies and a real depreciation of currencies with respect to the US dollar. We find considerable evidence for heterogeneity in the spillovers across countries, as well for changes in the transmission of monetary policy shocks over time. (authors' abstract) / Series: Department of Economics Working Paper Series

This article assesses the empirical relationship between per capita income growth fluctuations and the age-structured human capital variations across four groups of geographically clustered developed and developing countries from spatial perspective. We estimate a spatial Vector Autoregressive (VAR) model of income dynamics where the distance between countries is defined on relational space based on their similarity in appropriation tendency of human capital in the production processes. These distances are computed using a newly developed human capital data set which fully characterizes the demographic structure of human capital, and thus underlines the joint relevance of demography and human capital in economic growth. Spatial effects on growth interdependence and complementarity are then explored with respect to the proposed distance metrics. Our results imply that significant cross-country growth interdependence based on human capital distances exists among defined country groups suggesting the need for a cooperative policy programme among them. We also find that the relationship between economic growth and human capital is highly nonlinear as a function of the proposed human capital distance.

This study was issued by Swedbank because they wanted too improve their GDP growth forecast capabilites.  A program was developed and tested on six countries; USA, Sweden, Germany, UK, Brazil and Norway. In this paper I investigate if I can reduce forecasting error for GDP growth by taking a smart average from a variety of models compared to both the best individual models and a random walk. I combine the forecasts from four model groups: Vector autoregression, principal component analysis, machine learning and random walk. The smart average is given by a system that give more weight to the predictions of models with a lower historical error. Different weighting schemas are explored; how far into the past should we look? How much should bad performance be punished? I show that for the six countries studied the smart average outperforms the single best model and that for five out of six countries it beats a random walk by at least 25%. / Den här studien beställdes av Swedbank eftersom de ville förbättra sin BNP-prediktionsförmåga. Ett dataprogram utvecklades och testades på sex länder; USA, Sverige, Tyskland, Storbritannien, Brasilien och Norge. I den här rapporten undersöker jag om jag kan minska felmarginalen för BNP-utvecklingsprognoser genom att ta ett smart genomsnitt från flera olika modeller jämfört med både den bästa individuella modellen och en random walk. Jag kombinerar prognoser från fyra modellgrupper: Vektor autoregression, principalkomponentanalys, maskininlärning och random walk. Det smarta genomsnittet skapas genom att ge mer vikt till de modeller som har lägst historiskt felmarginal. Olika viktningsscheman utforskas; hur långt bak i tiden ska vi mäta? Hur hårt ska dåliga prediktioner bestraffas? Jag visar att för de sex länderna i studien presterar det smarta genomsnittet bättre än den enskilt bästa modellen och fem av de sex länderna slår en random walk med mer än 25%.

This study constructs a financial stress index with a specific focus on the case of the Czech Republic. The advantage of the index is primarily its ability to measure the current level of stress in the financial system incorporating information from various sectors of the economy and expressing it in a single-value statistic. Our index successfully recorded and evaluated critical periods of elevated financial stress especially during the recent financial crisis. Furthermore, we examine a systematic interaction between financial stress and the macroeconomics using vector autoregression analysis along with method of impulse responses. Based on our results we observe a significant and positive response of unemployment due to the shock in financial stress. Conversely, a negative effect was examined on inflation and interest rates. JEL Classification G17, G32 Keywords financial stress index, vector autoregression, impulse responses

Le ciblage de l’inflation est le cadre le plus récent de la politique monétaire dans le monde. Il est désormais largement choisi par les pays avancés ainsi que par les pays émergents. Cependant, deux questions principales sont encore en débat particulièrement dans les pays émergent et en développement. Ils s’agissent des avantages du ciblage d’inflation et du respect de conditions préalables afin d’assurer le succès de ce régime. Empiriquement, on conclut que le ciblage d’inflation est un cadre de politique monétaire réussie pour les pays émergents. En plus, il n'est pas nécessaire pour ces pays de satisfaire toutes les conditions préalables strictes avant de réussir à l'adopter. La situation budgétaire et l'indépendance de la banque centrale jouent un rôle plus important que les autres conditions et doivent être préparées en premier lieu. Concernant le Vietnam, par l'approche structurelle vecteur autorégressif (VAR), la thèse montre que la politique monétaire de la banque centrale n’est pas efficace. Donc, il permet de confirmer la nécessité du changement de stratégie monétaire par rapport au cadre actuel. Cependant, notamment parce que la banque centrale n’est pas indépendante, le Vietnam ne peut pas adopter le ciblage d’inflation dans un bref délai. Les recommandations du durcissement de la contrainte budgétaire et de l’augmentation l’indépendance de la banque centrale sont suggérées avant la mise en œuvre de sa stratégie de ciblage d’inflation. / Inflation targeting (hereafter IT) is the newest monetary policy framework in the world. The practice of IT has been chosen by both advanced countries and emerging countries. However, two main issues are still under debate particularly in emerging and developing countries. They are the benefits of IT and preconditions to success adoption. Empirically, we showed that IT is considered as a successful monetary policy framework for emerging countries. In addition, it is not necessary for emerging markets to satisfy all stringent preconditions to successfully adopt IT. In practice, the fiscal situation and the central bank independence play a more important role than other conditions and need to be prepared first.Basing on Structural Vector Autoregression (SVAR), the thesis concludes that Vietnamese monetary policy currently does not effectively control the inflation rate. Inflation targeting framework would be a solution to this. Nonetheless, this thesis concludes that at this moment in time Vietnam is not able to adopt the IT framework, as it still must prepare some of the preconditions required before official adoption. The recommendations of hardening the budget constraint and increase central bank independence in relationship with government are suggested before implementing IT strategy.

This work combines discrete and continuous methods while modeling connect- edness of financial tick data. As discrete method we are using vector autore- gression. For continuous domain Hawkes process is used, which is special case of point process. We found out that financial assets are connected in non- symmetrical fashion. By using two methodologies we were able to model bet- ter how are the series connected. We confirmed existence of price leader in our three stock portfolio and modeled connectedness of jumps between stocks. As conclusion we state that both methods yields important results about price nature on the market and should be used together or at least with awareness of second approach. JEL Classification C32, G11, G14 Keywords Vector Autoregression, Hawkes process, High- frequency analysis, Connectedness Author's e-mail petr.petras@email.cz Supervisor's e-mail krehlik@utia.cas.cz

Digitized acoustical signals of Byzantine music performed by Iakovos Nafpliotis are used to extract the fundamental frequency of each note of the diatonic scale. These empirical results are then contrasted to the theoretical suggestions and previous empirical findings. Several parametric and non-parametric spectral parameter estimation methods are implemented. These include: (1) Phase vocoder method, (2) McAulay-Quatieri method, (3) Levinson-Durbin algorithm,(4) YIN, (5) Quinn & Fernandes Estimator, (6) Pisarenko Frequency Estimator, (7) MUltiple SIgnal Characterization (MUSIC) algorithm, (8) Periodogram method, (9) Quinn & Fernandes Filtered Periodogram, (10) Rife & Vincent Estimator, and (11) the Fourier transform. Algorithm performance was very precise. The psychophysical aspect of human pitch discrimination is explored. The results of eight (8) psychoacoustical experiments were used to determine the aural just noticeable difference (jnd) in pitch and deduce patterns utilized to customize acceptable performable pitch deviation to the application at hand. These customizations [Acceptable Performance Difference (a new measure of frequency differential acceptability), Perceptual Confidence Intervals (a new concept of confidence intervals based on psychophysical experiment rather than statistics of performance data), and one based purely on music-theoretical asymphony] are proposed, discussed, and used in interpretation of results. The results suggest that Nafpliotis' intervals are closer to just intonation than Byzantine theory (with minor exceptions), something not generally found in Thrasivoulos Stanitsas' data. Nafpliotis' perfect fifth is identical to the just intonation, even though he overstretches his octaveby fifteen (15)cents. His perfect fourth is also more just, as opposed to Stanitsas' fourth which is directionally opposite. Stanitsas' tendency to exaggerate the major third interval A4-F4 is still seen in Nafpliotis, but curbed. This is the only noteworthy departure from just intonation, with Nafpliotis being exactly Chrysanthian (the most exaggerated theoretical suggestion of all) and Stanitsas overstretching it even more than Nafpliotis and Chrysanth. Nafpliotis ascends in the second tetrachord more robustly diatonically than Stanitsas. The results are reported and interpreted within the framework of Acceptable Performance Differences.

This thesis deals with INGARCH models for a count time series. Main emphasis is placed on a linear INARCH model. Its properties are derived. Several methods of estimation are introduced - maximum likelihood method, least squares method and its modifications - and later compared in a simulation study. Main properties and maximum likelihood estimation for INGARCH(1,1) model are stated. Higher order linear INGARCH models and nonlinear INGARCH models are discussed briefly. An application of the presented models on time series of car accidents is given.

A compreensão dos efeitos das atividades de marketing nas vendas de produtos de consumo em um ambiente multicanal é de fundamental importância para acadêmicos e executivos. As decisões sobre as variáveis controláveis de marketing para as marcas de consumo nos mercados considerados emergentes, como o Brasil, são desafiadoras porque, no contexto do varejo alimentar, os canais de distribuição nesses mercados variam mais em relação aos mercados maduros em termos de formatos e tipos de varejistas. No Brasil, o varejo é ainda menos concentrado do que em outros países desenvolvidos. Os supermercados de vizinhança e as lojas tradicionais independentes, como mercearias e padarias, ainda são importantes. Por outro lado, os grandes grupos varejistas têm expandido seus negócios. Considerando que esses canais variam em relação ao tamanho da loja e ao formato (autosserviço e full-service), variedade de SKUs (Stock Keeping Unit) oferecidos, propriedade e perfil do público-alvo, o efeito das atividades de marketing da indústria pode ser diferente em cada um desses canais. Nesse contexto, esta pesquisa investiga se os efeitos nas vendas provenientes das atividades de marketing, com foco em gerenciamento de canais e comunicação push (dirigidas aos canais) e pull (dirigidas aos consumidores finais), variam por canal de distribuição, mensurando, assim, quais são os efeitos nas vendas em cada canal. A base de dados utilizada estava disponível por SKU para todas as marcas de bebidas carbonatadas referentes a uma região do Brasil, que representa 16,5% das vendas no varejo alimentar. Os dados, no período de janeiro de 2008 até dezembro de 2011, estavam disponíveis mensalmente incluindo variáveis de produto, preço, cobertura de mercado e atividades promocionais para todos os SKUs de bebidas nos grandes supermercados (AS>5), pequenos supermercados (AS 1-4) e o canal formado pelas lojas full-service, (tradicional). Sobre os investimentos em comunicações de marketing, a base de dados foi disponibilizada por um fabricante de marcas líderes no mercado de bebidas. A metodologia deste trabalho, de abordagem quantitativa, envolveu os testes de validação e a aplicação do método de análise multivariada para séries temporais, seguindo o modelo de Vetores Autorregressivos (VAR). Um ponto de destaque desta pesquisa é a adaptação do modelo VAR para a modelagem das variáveis de marketing em um contexto multicanal, analisando os efeitos das atividades push e pull de maneira integrada com todas as variáveis controláveis de marketing (comunicação, preço, distribuição e produto). Mesmo pesquisas conduzidas em mercados maduros ainda não exploraram totalmente as sinergias entre as atividades push e pull em diferentes canais. Os resultados desta pesquisa indicaram que os efeitos das atividades de marketing variam por canal. As funções de respostas ao impulso, a partir das equações do modelo VAR, são apresentadas para cada atividade de marketing analisada, mensurando seu efeito nas vendas de cada canal. Isso permitiu analisar as hipóteses propostas. Por fim, este estudo contribui com uma metodologia que permite modelar as variáveis de marketing em um contexto multicanal e, ainda, apresenta o efeito das atividades de marketing nas vendas em cada tipo de varejista analisado. / Understanding marketing mix effects on consumer product\'s sales in a multichannel environment is of importance to both scholars and practitioners. Marketing mix decisions for consumer brands in emerging markets, such as Brazil, is challenging because in the grocery retailing, channels in these markets vary more than in the developed markets with regards to their format and type. In Brazil, the level of concentration in grocery retailing is still smaller in than in developed markets. Neighborhood stores and independent mom-and-pop stores are still of importance; however, big retailers\' chains are expanding their businesses. Considering that these channels vary in terms of store size (self-service and full-service), breadth of assortment, value proposition and customers\' profile, effects of manufacturers\' marketing activities might be different in each channel. Under this context, this research analyzes whether effects on sales from the marketing activities vary by channel, with focus on channel management and marketing push and pull. This assessment was possible by measuring what these effects are across channels. Data for the study comes from store audits that spans four years, from 2008 to 2011, for all brands in the carbonated soft-drinks category from a region in Brazil which accounts for 16,5% of sales in food retail. The data was available by channel and SKU, including channel management measures for all SKUs in big supermarkets (AS>5), small supermarkets (AS 1-4) and mom-and-pop stores. Data for the marketing communication spending came from a beverage leading company. The methodology used for this quantitative research included validation tests and the employment of a method for multivariate time series analysis, called Vector Autorregressive Models (VAR). A highlight of the study is the employment of a VAR model in a multichannel context, which makes it possible to analyze the effects of push and pull activities integrated with the others marketing variables (communication, price, distribution and product). Even research conducted in developed markets has not explored synergies between push and pull. Results from this research have indicated that the effects of marketing activities vary by channel. The impulse-reponse functions by each marketing activities and channels are estimated in order to test the hypothesis proposed in this study. Thus, it contributes to creating an understanding of how to model the marketing mix variables in a multichannel environment and to creating an understanding of what marketing activities are more potential to drive higher level of sales by each analyzed channel.