|
on Forecasting |
| By: | Catik, A. Nazif; Karaçuka, Mehmet |
| Abstract: | This paper analyses inflation forecasting power of artificial neural networks with alternative univariate time series models for Turkey. The forecasting accuracy of the models is compared in terms of both static and dynamic forecasts for the period between 1982:1 and 2009:12. We find that at earlier forecast horizons conventional models, especially ARFIMA and ARIMA, provide better one-step ahead forecasting performance. However, unobserved components model turns out to be the best performer in terms of dynamic forecasts. The superiority of the unobserved components model suggests that inflation in Turkey has time varying pattern and conventional models are not able to track underlying trend of inflation in the long run. -- |
| Keywords: | Inflation forecasting,Neural networks,Unobserved components model |
| JEL: | C45 C53 E31 E37 |
| Date: | 2011 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:dicedp:20&r=for |
| By: | Filip Novotny; Marie Rakova |
| Abstract: | Consensus Economics forecasts for euro-area GDP growth, consumer and producer price inflation and the USD/EUR exchange rate are used by the Czech National Bank to make assumptions about future external economic developments. This paper compares the accuracy of the aforementioned Consensus forecasts to those of the European Commission, International Monetary Fund and Organization for Economic Co-operation and Development, and also to the naïve forecast and the forecast implied by the forward exchange rate. In the period from 1994 to 2009 Consensus forecasts for effective euro-area consumer price inflation and GDP growth beat the alternatives by a difference which is typically statistically significant. The results are more diverse for the pre-crisis sample (1994–2007). The Consensus forecast for euro-area producer price inflation significantly outperforms the naïve forecast in the short-term. Finally, the Consensus forecast for the USD/EUR exchange rate during the period from 2002 to 2009 is more precise than the naïve forecast and the forecast implied by the forward rate. |
| Keywords: | Forecasting accuracy, prediction process, survey forecasts. |
| JEL: | E37 E58 |
| Date: | 2010–12 |
| URL: | http://d.repec.org/n?u=RePEc:cnb:wpaper:2010/14&r=for |
| By: | Fossati, Sebastian (University of Alberta, Department of Economics) |
| Abstract: | A probit model is used to show that latent common factors estimated by principal components from a large number of macroeconomic time series have important predictive power for NBER recession dates. A pseudo out-of-sample forecasting exercise shows that predicted recession probabilities consistently rise during subsequently declared NBER recession dates. The latent variable in the factor-augmented probit model is interpreted as an index of real business conditions which can be used to assess the strength of an expansion or the depth of a recession. |
| Keywords: | business cycle; forecasting; factors; probit model; Bayesian methods |
| JEL: | C01 C22 C25 E32 E37 |
| Date: | 2011–05–01 |
| URL: | http://d.repec.org/n?u=RePEc:ris:albaec:2011_005&r=for |
| By: | Marcus Cobb; Gonzalo Echavarría; Pablo Filippi; Macarena García; Carolina Godoy; Wildo González; Carlos Medel; Marcela Urrutia |
| Abstract: | The aim of this document is to provide a forecasting tool that facilitates understanding economic developments in a timely manner. This is pursued through the Bridge Model approach by using it to relate a large set of monthly indicators to Chilean GDP and its main components. The outcome is a set of simple equations that characterize reasonably well total GDP and the feasible supply- and demand-side components based on a small set of relevant indicators. The selected equations generally provide better short-term forecasts than simple autoregressive models. However, if needed, the equation selection methodology is straightforward enough to update the equations easily making it an attractive tool for real-time forecasting. |
| Date: | 2011–05 |
| URL: | http://d.repec.org/n?u=RePEc:chb:bcchwp:626&r=for |
| By: | Jan Filacek; Branislav Saxa |
| Abstract: | Do private analysts coordinate their forecasts via central bank forecasts? In this paper, we examine private and central bank forecasts for the Czech Republic. The evolution of the standard deviation of private forecasts as well as the distance from the central bank’s forecasts are used to study whether a coordination effect exists, how it is influenced by uncertainty, and the effects of changes in central bank communication. The results suggest that private analysts coordinate their forecasts for the interest rate and inflation, while no or limited evidence exists for the exchange rate and GDP growth. |
| Keywords: | Central bank, coordination, forecast. |
| JEL: | E27 E37 E47 E58 |
| Date: | 2010–12 |
| URL: | http://d.repec.org/n?u=RePEc:cnb:wpaper:2010/13&r=for |
| By: | Prof. Dr. Abu Hassan Shaari Md Nor; Behrooz Gharleghi (Faculty of Economic and Management, Universiti Kebangsaan Malaysia); Prof. Dr. Khairuddin Omar (Faculty of Science Information and Technology, Universiti Kebangsaan Malaysia); Dr. Tamat Sarmidi (Faculty of Economic and Management, Universiti Kebangsaan Malaysia) |
| Abstract: | Financial systems, such as the exchange rate market, are generally complex systems with limited information about the underlying mechanisms governing the data. Such systems are often characterised as ‘‘black boxes” and hence the internal mechanisms and relations among the elements of the system can be ignored while concentrating on the study of the relationship between the system’s input(s)/output(s). Financial systems are also generally assumed to be nonlinear systems, which increase the level of difficulty in accurately predicting their behaviour. Despite these difficulties, the financial implications of accurate prediction of the financial markets’ movements have encouraged researchers and practitioners to employ a variety of modelling methods |
| Keywords: | ANNs, ARIMA, hybrid models |
| JEL: | C22 C45 C53 |
| Date: | 2011–03 |
| URL: | http://d.repec.org/n?u=RePEc:cms:2icb11:2011-087&r=for |
| By: | Adi Schnytzer (Bar-Ilan University); Yisrael Schnytzer |
| Abstract: | Prediction markets have been shown to generate fairly accurate odds of various events occurring in the future. The forthcoming possibility of natural disasters provides, on occasion, an opportunity for a bet, yet no wide scale and accepted prediction market has arisen despite its obvious importance, probably due in part to its 'politically incorrect' nature, but more importantly to the fact that we have yet to develop accurate forecasting models. Animals, however, have been forced through natural selection to develop elaborate anticipatory mechanisms to predict possible upcoming calamites. Recent studies suggest that animals can, days and sometimes weeks in advance, predict the occurrence of earthquakes. A wealth of recent observations and laboratory studies corroborate this. Natural disasters have lead to the development of 'risk taking' behaviors when the 'odds' of an upcoming disaster outweigh the benefits of maintaining territory, mating, and other basic behaviors. For various historical reasons discussed in this review, this field has had trouble coming of age, with little funding and support from the scientific community, particularly in the US. Due to the great odds at stake and the tremendous economic impact of earthquakes we wish to raise the awareness of this vital topic to the wider scientific community. We present a review of such animal predictive behavior and propose that an early “reading” of such models might lead to the development of a predictions market by humans. |
| Date: | 2011–05 |
| URL: | http://d.repec.org/n?u=RePEc:biu:wpaper:2011-20&r=for |
| By: | Imene Mootamri (GREQAM - Groupement de Recherche en Économie Quantitative d'Aix-Marseille - Université de la Méditerranée - Aix-Marseille II - Université Paul Cézanne - Aix-Marseille III - Ecole des Hautes Etudes en Sciences Sociales (EHESS) - CNRS : UMR6579) |
| Abstract: | The main purpose of this paper is to consider the multivariate GARCH (MGARCH) framework to model the volatility of a multivariate process exhibiting long term dependence in stock returns. More precisely, the long term dependence is examined in the first conditional moment of US stock returns through multivariate ARFIMA process and the time-varying feature of volatility is explained by MGARCH models. An empirical application to the returns series is carried out to illustrate the usefulness of our approach. The main results confi rm the presence of long memory property in the conditional mean of all stock returns. |
| Keywords: | Forecasting; Long memory; Multivariate GARCH; Stock Returns |
| Date: | 2011–06–09 |
| URL: | http://d.repec.org/n?u=RePEc:hal:wpaper:halshs-00599250&r=for |
| By: | Crescenzio Gallo; Michelangelo De Bonis |
| Abstract: | The photovoltaic (PV) industry in Italy has already crossed the threshold of 1 GW of installed capacity. Currently there are approximately 70,000 certified facilities in operation for a power generation of 1,300 GWh/year. With these figures, Italy has become the second country in Europe for PV installed power after Germany. The energy produced would be sufficient to meet the power needs of approximately 1,200,000 people. This leads to some questions: Will this technology continue to grow exponentially even after the recent reduction in rates by the Energy Bill? Will the number of installed PV facilities still grow even with less public support and (probably) a reduction in the technology purchase price? The purpose of this paper is therefore to develop a conceptual model to make a prediction of the PV installed power in Italy through the use of “supervised” artificial neural networks. This model is also applied to the analysis of the spread of this technology in some other European countries. |
| Keywords: | photovoltaic, forecasting, neural networks. |
| Date: | 2011–03 |
| URL: | http://d.repec.org/n?u=RePEc:ufg:qdsems:02-2011&r=for |
| By: | Saratha Sathasivam (School of Mathematical Sciences, Universiti Sains Malaysia) |
| Abstract: | Neural Networks represent a meaningfully different approach to using computers in the workplace. A neural network is used to learn patterns and relationships in data. The data may be the results of a market research effort, or the results of a production process given varying operational conditions. Regardless of the specifics involved, applying a neural network is a substantial departure from traditional approaches. In this paper we will look into how neural networks is used in data mining. The ultimate goal of data mining is prediction - and predictive data mining is the most common type of data mining and one that has the most direct business applications. Therefore, we will consider how this technique can be used to classify the performance status of a departmental store in monitoring their products |
| Keywords: | Neural networks, data mining, prediction |
| JEL: | M0 |
| Date: | 2011–03 |
| URL: | http://d.repec.org/n?u=RePEc:cms:2icb11:2011-137&r=for |
| By: | Menno Middeldorp |
| Abstract: | Central banks worldwide have become more transparent. An important reason is that democratic societies expect more openness from public institutions. Policymakers also see transparency as a way to improve the predictability of monetary policy, thereby lowering interest rate volatility and contributing to economic stability. Most empirical studies support this view. However, there are three reasons why more research is needed. First, some (mostly theoretical) work suggests that transparency has an adverse effect on predictability. Second, empirical studies have mostly focused on average predictability before and after specific reforms in a small set of advanced economies. Third, less is known about the effect on interest rate volatility. To extend the literature, I use the Dincer and Eichengreen (2007) transparency index for twenty-four economies of varying income and examine the impact of transparency on both predictability and market volatility. I find that higher transparency improves the accuracy of interest rate forecasts for three months ahead and reduces rate volatility. |
| Date: | 2011 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fednsr:496&r=for |
| By: | Stéphane Dées (European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.); Pedro Soares Brinca (Stockholm University, Sweden.) |
| Abstract: | For most academics and policy makers, the depth of the 2007-09 financial crisis, its longevity and its impacts on the real economy resulted from an erosion of confidence. This paper proposes to assess empirically the link between consumer sentiment and consumption expenditures for the United States and the euro area. It shows under which circumstances confidence indicators can be a good predictor of household consumption even after controlling for information in economic fundamentals. Overall, the results show that the consumer confidence index can be in certain circumstances a good predictor of consumption. In particular, out-of-sample evidence shows that the contribution of confidence in explaining consumption expenditures increases when household survey indicators feature large changes, so that confidence indicators can have some increasing predictive power during such episodes. Moreover, there is some evidence of a "confidence channel" in the international transmission of shocks, as U.S. confidence indices lead consumer sentiment in the euro area. JEL Classification: C32, E17, F37, F42. |
| Keywords: | Consumer Confidence, Consumption, International Linkages, Non-linear modeling. |
| Date: | 2011–06 |
| URL: | http://d.repec.org/n?u=RePEc:ecb:ecbwps:20111349&r=for |