<p>In this paper, the forecast performance of an unrestricted Vector Autoregressive (VAR) model was compared against the forecast accuracy of a Vector error correction (VECM) model when computing out-of-sample forecasts for Swedish exports. The co-integrating relation used to estimate the error correction specification was based upon an economic theory for international trade suggesting that a long run equilibrium relation among the variables included in an export demand equation should exist. The results obtained provide evidence of a long run equilibrium relationship between the Swedish export volume and its main determinants. The models were estimated for manufactured goods using quarterly data for the period 1975-1999 and once estimated, the models were used to compute out-of-sample forecasts up to four-, eight- and twelve-quarters ahead for the Swedish export volume using both multi-step and one-step ahead forecast techniques. The main results suggest that the differences in forecasting ability between the two models are small, however according to the relevant evaluation criteria the unrestricted VAR model in general yields somewhat better forecast than the VECM model when forecasting Swedish exports over the chosen forecast horizons.</p>

In this paper, the forecast performance of an unrestricted Vector Autoregressive (VAR) model was compared against the forecast accuracy of a Vector error correction (VECM) model when computing out-of-sample forecasts for Swedish exports. The co-integrating relation used to estimate the error correction specification was based upon an economic theory for international trade suggesting that a long run equilibrium relation among the variables included in an export demand equation should exist. The results obtained provide evidence of a long run equilibrium relationship between the Swedish export volume and its main determinants. The models were estimated for manufactured goods using quarterly data for the period 1975-1999 and once estimated, the models were used to compute out-of-sample forecasts up to four-, eight- and twelve-quarters ahead for the Swedish export volume using both multi-step and one-step ahead forecast techniques. The main results suggest that the differences in forecasting ability between the two models are small, however according to the relevant evaluation criteria the unrestricted VAR model in general yields somewhat better forecast than the VECM model when forecasting Swedish exports over the chosen forecast horizons.

<p><strong>Syfte:</strong> Syftet är att försöka urskilja samband mellan hushålls socioekonomiska faktorer och efterfrågan på specifika bostadsattribut. Vår frågeställning är: Finns det samband mellan vilka specifika attribut som efterfrågas hos bostadsrätter och hushålls socioekonomiska faktorer? Vår förhoppning är att åtminstone få en indikation på vad olika typer av hushåll efterfrågar.<strong>Metod: </strong>Denna studie baseras på en databas bestående av bostadsrättsförsäljningar som skett i Gävle under år 2008 samt socioekonomisk information om de hushåll som förvärvat dessa bostadsrätter. För att estimera hushålls efterfrågan på ett specifikt bostadsrättsattribut har en tvåstegsmetod använts. I ett första steg avslöjas de underförstådda marginalpriserna av bostadsrätters egenskaper med hjälp av den hedoniska metoden. Dessa marginalpriser används i ett andra steg där efterfrågeekvationer för enskilda bostadsrättsattribut estimeras. Beräkningarna utförs i statistikprogrammet EViews.<strong>Resultat & slutsats: </strong>Resultatet visar att vissa bostadsrättsattribut kan sammankopplas med hushålls socioekonomiska faktorer. Vidare visar resultatet tydliga men mycket svaga mönster vad gäller efterfrågan på marknaden, trots att prisstrukturen på samma marknad är oerhört utmärkande. Vi kan konstatera att prissättningen på bostadsrätters attribut är väldigt tydlig medan konsumtionen av samma attribut är otydlig.<strong>Förslag till fortsatt forskning:</strong> Intressant vore att jämföra studier av denna typ med hyresmarknadens hyressättningsmodell och se om den utgår från samma värdering av bostadens attribut som det visat sig att hushåll efterfrågar. Ytterligare ett förslag är att skatta en faktisk boendekostnad för samtliga hushåll i datamaterialet och använda disponibel inkomst istället för taxerad förvärvsinkomst i efterfrågefunktionen.<strong>Uppsatsens bidrag: </strong>Studien har visat hur man kan estimera och finna samband mellan bostadsattribut och hushålls socioekonomiska faktorer. Därutöver har problemet med endogenitet behandlats genom instrumentvariabler och en uppdelning av datamaterialet i fyra delområden.</p> / <p><strong>Aim:</strong> The purpose of this study is to try to distinguish a relationship between household socio-economic factors and the demand for specific housing attributes. Our question is: Is there a connection between the specific attributes requested by households and its socio-economic factors? Our hope is to at least get an indication of what different types of households demand.<strong>Method:</strong> This study is based on information of tenant-owner flats sales made in Gävle in 2008 as well as socio-economic information of those households who bought these flats. In order to estimate demand for a specific housing attribute a two-step method is used. In a first step the implicit marginal prices of housing attributes are revealed by the hedonic method. These marginal rates are used in a second step to reveal the specific household demand for individual housing attributes. The calculations are made in the statistical program EViews.<strong>Result & Conclusions: </strong>The result shows that some housing attributes can be linked with household socio-economic factors. The result shows a clear but very weak pattern of demand in the market, despite that the price structure in the same market is extremely remarkable. We note that the prices of housing attributes are very clear while the consumption of the same attributes is unclear.<strong>Suggestions for future research: </strong>It would be interesting to compare the results of this type of study with rental markets rent-model and see if it is based on the same valuation of the dwelling attributes that household’s demand. Another proposal is to estimate the actual housing costs for all of the households in the data and use disposable income rather than actual income in the demand function.<strong>Contribution of the thesis: </strong>The study has revealed how to estimate and identify household demand for specific housing attributes. In addition, the problem of endogeneity has been treated with instrument variables and a separation of the data set into four submarkets.</p>

Syfte: Syftet är att försöka urskilja samband mellan hushålls socioekonomiska faktorer och efterfrågan på specifika bostadsattribut. Vår frågeställning är: Finns det samband mellan vilka specifika attribut som efterfrågas hos bostadsrätter och hushålls socioekonomiska faktorer? Vår förhoppning är att åtminstone få en indikation på vad olika typer av hushåll efterfrågar.Metod: Denna studie baseras på en databas bestående av bostadsrättsförsäljningar som skett i Gävle under år 2008 samt socioekonomisk information om de hushåll som förvärvat dessa bostadsrätter. För att estimera hushålls efterfrågan på ett specifikt bostadsrättsattribut har en tvåstegsmetod använts. I ett första steg avslöjas de underförstådda marginalpriserna av bostadsrätters egenskaper med hjälp av den hedoniska metoden. Dessa marginalpriser används i ett andra steg där efterfrågeekvationer för enskilda bostadsrättsattribut estimeras. Beräkningarna utförs i statistikprogrammet EViews.Resultat &amp; slutsats: Resultatet visar att vissa bostadsrättsattribut kan sammankopplas med hushålls socioekonomiska faktorer. Vidare visar resultatet tydliga men mycket svaga mönster vad gäller efterfrågan på marknaden, trots att prisstrukturen på samma marknad är oerhört utmärkande. Vi kan konstatera att prissättningen på bostadsrätters attribut är väldigt tydlig medan konsumtionen av samma attribut är otydlig.Förslag till fortsatt forskning: Intressant vore att jämföra studier av denna typ med hyresmarknadens hyressättningsmodell och se om den utgår från samma värdering av bostadens attribut som det visat sig att hushåll efterfrågar. Ytterligare ett förslag är att skatta en faktisk boendekostnad för samtliga hushåll i datamaterialet och använda disponibel inkomst istället för taxerad förvärvsinkomst i efterfrågefunktionen.Uppsatsens bidrag: Studien har visat hur man kan estimera och finna samband mellan bostadsattribut och hushålls socioekonomiska faktorer. Därutöver har problemet med endogenitet behandlats genom instrumentvariabler och en uppdelning av datamaterialet i fyra delområden. / Aim: The purpose of this study is to try to distinguish a relationship between household socio-economic factors and the demand for specific housing attributes. Our question is: Is there a connection between the specific attributes requested by households and its socio-economic factors? Our hope is to at least get an indication of what different types of households demand.Method: This study is based on information of tenant-owner flats sales made in Gävle in 2008 as well as socio-economic information of those households who bought these flats. In order to estimate demand for a specific housing attribute a two-step method is used. In a first step the implicit marginal prices of housing attributes are revealed by the hedonic method. These marginal rates are used in a second step to reveal the specific household demand for individual housing attributes. The calculations are made in the statistical program EViews.Result &amp; Conclusions: The result shows that some housing attributes can be linked with household socio-economic factors. The result shows a clear but very weak pattern of demand in the market, despite that the price structure in the same market is extremely remarkable. We note that the prices of housing attributes are very clear while the consumption of the same attributes is unclear.Suggestions for future research: It would be interesting to compare the results of this type of study with rental markets rent-model and see if it is based on the same valuation of the dwelling attributes that household’s demand. Another proposal is to estimate the actual housing costs for all of the households in the data and use disposable income rather than actual income in the demand function.Contribution of the thesis: The study has revealed how to estimate and identify household demand for specific housing attributes. In addition, the problem of endogeneity has been treated with instrument variables and a separation of the data set into four submarkets.

Made available in DSpace on 2015-03-05T18:58:03Z (GMT). No. of bitstreams: 0 Previous issue date: 8 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Este estudo tem por objetivo a estimação de equações de importação brasileira para bens de consumo duráveis, bens intermediários e bens de capital no período entre os anos 1988 a 2008. Além de realizar revisão sobre as teorias do comércio internacional, destacando os argumentos favoráveis ao livre comércio e abordando a nova teoria do comércio internacional, analisam-se a evolução das importações brasileiras com ênfase ao processo de abertura comercial. A abertura comercial brasileira possibilitou um aumento do quantum importado beneficiando o setor produtivo, pois a participação no ano de 2008 dos bens de capital e bens intermediários nas importações foi 58% e 14%, respectivamente. Estimam-se funções de longo prazo, com os ajustes de curto prazo estimados pelo Mecanismo de Correção de Erros (MCE). Os vetores de co-integração são estimados através de procedimentos de Johansen. As importações das três categorias de uso nos modelos de longo prazo revelaram-se altamente dependentes das oscilações da renda brasil / This study aims to estimate equations for Brazilian imports of durable consumer goods, intermediate goods and capital goods in the period from 1988 to 2008. In addition to reviewing the theories of international trade, highlighting the arguments in favor of free trade and addressing the new international trade theory, analyzes the evolution of Brazilian imports with emphasis on trade liberalization. The Brazilian trade openness has enabled an increase in the volume of imports benefiting most the productive sector, for participation in 2008 of capital goods and intermediate goods in imports was 58% and 14% respectively. Estimated functions of long-term with short-term adjustments estimated by Errors of Correction Mechanism (ECM). The vector co-integration is estimated by procedure of Johansen. Imports of the three categories of use in models of long-term proved to be highly dependent on fluctuations in income in Brazil. For the income coefficient 5.17 was obtained in capital goods, intermediate goods at 4.63 a

Submitted by Cristiane Shirayama (cristiane.shirayama@fgv.br) on 2011-06-03T17:27:31Z No. of bitstreams: 1 66080100267.pdf: 616070 bytes, checksum: 5cacc6bc2a5f7bf30b0b0e4c76930811 (MD5) / Approved for entry into archive by Gisele Isaura Hannickel(gisele.hannickel@fgv.br) on 2011-06-03T17:34:19Z (GMT) No. of bitstreams: 1 66080100267.pdf: 616070 bytes, checksum: 5cacc6bc2a5f7bf30b0b0e4c76930811 (MD5) / Approved for entry into archive by Suzinei Teles Garcia Garcia(suzinei.garcia@fgv.br) on 2011-06-03T17:35:34Z (GMT) No. of bitstreams: 1 66080100267.pdf: 616070 bytes, checksum: 5cacc6bc2a5f7bf30b0b0e4c76930811 (MD5) / Made available in DSpace on 2011-06-03T18:07:37Z (GMT). No. of bitstreams: 1 66080100267.pdf: 616070 bytes, checksum: 5cacc6bc2a5f7bf30b0b0e4c76930811 (MD5) Previous issue date: 2011-02-11 / O trabalho estima um sistema de equações de demanda para os produtos do mercado de frios no Brasil, utilizando o modelo Almost Ideal Demand System (AIDS) em três estágios. O passo seguinte à estimação das equações de demanda é a avaliação das elasticidades própria e cruzada obtidas, verificando se estas são condizentes com as previsões da teoria econômica, o que permitiria a utilização destas para simular os efeitos de uma fusão entre as empresas Sadia e Perdigão no mercado de frios. Apenas os resultados dos estágios superior e intermediário – no qual são estimadas as elasticidades relativas aos diferentes tipos de frios, como presunto e salame – estão de acordo com o que seria previsto pela teoria. Já no estágio inferior, onde se estimam as elasticidades relativas às diferentes marcas para cada tipo de frio, isso não acontece. Uma provável causa dessa inconsistência nos resultados está relacionada ao problema de endogeneidade inerente à estimação de equações de demanda com dados agregados, uma vez que, para o estágio inferior, não foi possível obter variáveis instrumentais suficientemente correlacionadas com as variáveis endógenas. Dados os resultados insatisfatórios para o estágio inferior, a simulação de fusão não foi feita. O trabalho apenas discute as inevitáveis inconsistências que ocorreriam na simulação de fusão em decorrência do uso de resultados incoerentes do ponto de vista teórico. / This work estimates a system of demand equations for the Brazilian processed meat products, using the Almost Ideal Demand System (AIDS) in three stages. The next step after the demand equation estimation is to evaluate the own and cross elasticities, verifying if its values is consistent with the economic theory, which would allow to use them to simulate the effects of the merge between Sadia and Perdigão. Only in the top and middle level equations – where the elasticities relative to the processed meat products such as ham and salami are estimated – the results are consistent with the economic theory. However, in the bottom level, where the elasticities among the different brands in each product are calculated, this doesn’t happen. One possible cause of this inconsistency in the outputs may be related to the endogeinity inherent in the demand equations estimation with aggregated data, once in the bottom level it was not possible to get instrumental variables sufficiently correlated with the endogenous variables. Due to the unsatisfying outputs of the bottom level, the merger simulation was not done. The work just discuss the inevitable inconsistencies that would occur in the merge simulation caused by the incoherent results.