Spelling suggestions: "subject:"[een] NOWCAST"" "subject:"[enn] NOWCAST""
1 |
NOWCASTING THE SWEDISH UNEMPLOYMENT RATE USING GOOGLE SEARCH DATAInganäs, Jacob January 2023 (has links)
In this thesis, the usefulness of search engine data to nowcast the unemployment rate of Sweden is evaluated. Four different indices from Google Trends based on keywords related to unemployment are used in the analysis and six different regARIMA models are estimated and evaluated. The results indicate that the fit is improved for models when data from Google Trends is included. To evaluate the nowcast ability of models, one-step-ahead predictions are calculated. Although the prediction error is lower for the models with data from Google Trends, Diebold-Mariano tests do not indicate that the predictions are significantly better compared topredictions from a model without data from Google Trends. It is therefore concluded that one cannot state that data from Google Trends improves nowcasts of the unemployment rate of Sweden. Additionally, predictions are calculated for longer forecast horizons. This analysis indicates that Google search data could be useful to forecast the unemployment rate of longerforecast horizons.
|
2 |
[pt] ESTIMANDO NOWCASTS PARA O PIB E INFLAÇÃO BRASILEIRA: UMA ABORDAGEM DE ESTADO-ESPAÇO APLICADA AO MODELO DE FATORES / [en] NOWCASTING BRAZILIAN GDP AND INFLATION: A STATE-SPACE APPOACH FOR FACTOR MODELSSAVIO CESCON GOULART BARBOSA 04 February 2020 (has links)
[pt] Nesse artigo aplicamos a técnica de estimação dos nowcasts apresentada por Giannone, Reichlin e Small (2008), para o PIB e inflação brasileiros. Extraímos informações de um elevado número de variáveis e produzimos modelos capazes de informar contemporaneamente uma medida para as variáveis em questão. Em posse dessa leitura cotidiana, produzida por esses modelos, estimamos uma regra de Taylor diária para o Banco Central do Brasil (BCB), o que permitiu melhor identificar choques monetários e alterações na função de reação do BCB ao longo do tempo. Concluímos, primeiramente, que os modelos nowcasts apresentam acurácia comparável às previsões do relatório Focus do BCB. Segundo, 2 (duas) comparações históricas realizadas mostraram indícios que nossa proxy para choques monetários diários está relacionada às decisões explícitas de política monetária. Por fim, encontramos evidências que os modelos nowcasts puderam capturar grande parte da informação relevante para a determinação da taxa de juros de curto prazo, o que deveria estimular a aplicação de tais modelos nos processos decisórios públicos e privados. / [en] In this article we apply the two-steps nowcasting method, described in Giannone, Reichlin, and Small (2008), to build nowcast models for Brazilian GDP and inflation. Throught the application of this method, we could extract information from a large data-set and build models which could be used to produce a daily measurement of GDP and inflation. Using this measurement was possible to build a daily Taylor rule for the Brazilian Central Bank (BCB). This new application of nowcast models allowed us to extract a daily measurement of monetary shocks. Our study produced three main findings. First, the nowcast model showed an accuracy close to projections presented in the Focus survey. Second, we identified by historical comparison that the monetary shocks proxy, measured by the differences between the daily Taylor rule and the movements in the short-term interest rate, are related with unanticipated monetary policies decisions. Finally, nowcasts were able to capture a great part of relevant information to determine the short-term interest rate, which should stimulate the policymakers and financial markets members to apply those models.
|
3 |
GDP forecasting and nowcasting : Utilizing a system for averaging models to improve GDP predictions for six countries around the worldLundberg, Otto January 2017 (has links)
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%.
|
4 |
Indicador coincidente da atividade econômica: uma aplicação à economia brasileiraPimentel, Luana Moreira de Miranda 19 February 2018 (has links)
Submitted by Luana Moreira de Miranda Pimentel (luana.miranda@hotmail.com) on 2018-03-06T15:26:03Z
No. of bitstreams: 1
DissertaçãoFinal_Luana Miranda.pdf: 878923 bytes, checksum: 0cdfc20956cc28d42465a96408817d60 (MD5) / Approved for entry into archive by GILSON ROCHA MIRANDA (gilson.miranda@fgv.br) on 2018-03-12T19:37:38Z (GMT) No. of bitstreams: 1
DissertaçãoFinal_Luana Miranda.pdf: 878923 bytes, checksum: 0cdfc20956cc28d42465a96408817d60 (MD5) / Made available in DSpace on 2018-03-16T16:57:35Z (GMT). No. of bitstreams: 1
DissertaçãoFinal_Luana Miranda.pdf: 878923 bytes, checksum: 0cdfc20956cc28d42465a96408817d60 (MD5)
Previous issue date: 2018-02-19 / Neste trabalho, desenvolvemos um indicador coincidente da atividade econômica brasileira de frequência mensal, construído a partir de um modelo na forma espaço de estados e estimado por meio do filtro de Kalman. O modelo impõe uma restrição de que a soma da taxa instantânea de crescimento mensal do nosso indicador nos meses correspondentes ao trimestre deve se igualar à taxa instantânea de crescimento trimestral do PIB observado, divulgado pelo IBGE. Essa disciplina imposta pelo modelo e a escolha minuciosa das variáveis auxiliares, permitiu a obtenção de resultados satisfatórios nas análises dentro e fora da amostra. As variáveis auxiliares foram selecionadas com base em critérios estatísticos e também tendo em vista um modelo estrutural, que se fundamenta na Equação de Apreçamento dos Ativos e relaciona o spread atual entre retornos de ativos de maturidades distintas com a taxa de crescimento da atividade econômica futura. O indicador construído é útil para prever o crescimento do PIB tanto em trimestres passados que ainda não foram divulgados (backcast) quanto no trimestre corrente (nowcast).
|
5 |
Essays on Forecasting Methods and Monetary Policy EvaluationLópez Buenache, Germán 27 July 2015 (has links)
No description available.
|
6 |
Examination of the Barotropic Behavior of the Princeton Coastal Ocean Model in Lake Erie, Using Water Elevations From Gage Stations and Topex/Poseidon AltimetersVelissariou, Vasilia 30 September 2009 (has links)
No description available.
|
Page generated in 0.0498 seconds