|
on Forecasting |
| By: | Wolfgang Härdle; Alena Mysickova |
| Abstract: | Population forecasts are crucial for many social, political and economic decisions. Official population projections rely in general on deterministic models which use different scenarios for future vital rates to indicate uncertainty. However, this technique shows substantial weak points such as assuming absolute correlations between the demographic components. In this paper, we argue that a stochastic projection alternative, with no a priori assumptions provides point forecasts and probabilistic prediction intervals for demographic parameters in addition. Age-sex specific population forecast for Germany is derived through a stochastic population renewal process using forecasts of mortality, fertility and migration. Time series models with demographic restrictions are used to describe immigration, emigration and time varying indices of mortality and fertility rates. These models are then used in the simulation of future vital rates to obtain age-specific population forecast using the cohort-component method. The consequence for the German pension system is discussed. To maintain the actual average pension level the premium rate of the present system rises at least by 50% as the old-age ratio nearly doubles by 2040. |
| Keywords: | Demographic Forecasting, Population Projection, Stochastic Demography |
| JEL: | J11 J13 C53 C22 |
| Date: | 2009–02 |
| URL: | http://d.repec.org/n?u=RePEc:hum:wpaper:sfb649dp2009-009&r=for |
| By: | Alessandra Amendola; Giuseppe Storti |
| Abstract: | This paper proposes a novel approach to the combination of conditional covariance matrix forecasts based on the use of the Generalized Method of Moments (GMM). It is shown how the procedure can be generalized to deal with large dimensional systems by means of a two-step strategy. The finite sample properties of the GMM estimator of the combination weights are investigated by Monte Carlo simulations. Finally, in order to give an appraisal of the economic implications of the combined volatility predictor, the results of an application to tactical asset allocation are presented. |
| Keywords: | Multivariate GARCH, Forecast Combination, GMM, Portfolio Optimization |
| JEL: | C52 C53 C32 G11 |
| Date: | 2009–01 |
| URL: | http://d.repec.org/n?u=RePEc:hum:wpaper:sfb649dp2009-007&r=for |
| By: | Reyes, Celia M; Domingo, Sonny N.; Gonzales, Kathrina G.; Mina, Christian D. |
| Abstract: | <p>Seasonal climate forecast (SCF) is one of the tools that could help farmers and decisionmakers better prepare for seasonal variability. However, a cloud of uncertainty looms over the true value of SCF to its target users. To shed light on the true value of SCF in local agricultural decisionmaking and operations, farm and household level survey was conducted. A total of 85 corn farmers from the plains and highlands of Echague and Angadanan, Isabela were interviewed.</p> <p>Results showed that climate and climate-related information were undoubtedly among the major factors being considered by farmers in their crop production activities. All aspects explored on the psychology of corn growers pointed to the high level of importance given to climatic conditions and SCF use. This was evident on the farmers’ perceptions, attitudes, and decisionmaking processes. Though the high regard of farmers on climate forecast and information cannot be questioned, actual application of such information seemed still wanting. Reliable indigenous knowledge on climate forecasting was scarce. With corn farmers in Isabela still thirsting for climate-related information, the delivery of appropriate information and accurate forecasts should be addressed through proper extension and provision of support.</p> |
| Keywords: | seasonal climate forecast (SCF), corn productivity, Isabela corn industry, climate variability, climate information and corn farming |
| Date: | 2009 |
| URL: | http://d.repec.org/n?u=RePEc:phd:dpaper:dp_2009-06&r=for |
| By: | Bergman, Ann (Arbetsvetenskap, Karlstads universitet); Karlsson, Jan Ch (Arbetsvetenskap, Karlstads universitet) |
| Abstract: | The article presents a typology of forecasts, i.e. statements on future events or states. It has two dimensions, truth claim and explanatory claim; each dimension has two values, making the claim or not making the claim. The four outcomes are then: Forecasts which make both truth claims and explanatory claims (predictions); which make truth claims, but not ex¬planatory claims (prognoses); which make explanatory claims, but not truth claims (science fiction); and which make neither truth claims nor explana¬tory claims (utopias or dystopias). We regard each outcome as an ideal type, against which forecasts can be measured. We illustrate the use of the typol¬ogy by presenting a Swedish example of each outcome. The prediction is Dahlbom’s The future of Sweden (Sveriges framtid), the prognosis is Wahl¬ström’s After the acid test (Efter stålbadet), the science fiction contribution is Martinson’s Aniara, and the dystopia is Boye’s Kallocain. |
| Keywords: | typology of forecasts; forecasts |
| JEL: | Z00 |
| Date: | 2009–03 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:ifswps:2009_003&r=for |
| By: | Drew Creal; Siem Jan Koopman; Andre Lucas |
| Abstract: | We propose a new class of observation driven time series models that we refer to as Generalized Autoregressive Score (GAS) models. The driving mechanism of the GAS model is the scaled likelihood score. This provides a unified and consistent framework for introducing time-varying parameters in a wide class of non-linear models. The GAS model encompasses other well-known models such as the generalized autoregressive conditional heteroskedasticity, autoregressive conditional duration, autoregressive conditional intensity and single source of error models. In addition, the GAS specification gives rise to a wide range of new observation driven models. Examples include non-linear regression models with time-varying parameters, observation driven analogues of unobserved components time series models, multivariate point process models with time-varying parameters and pooling restrictions, new models for time-varying copula functions and models for time-varying higher order moments. We study the properties of GAS models and provide several non-trivial examples of their application. |
| Keywords: | dynamic models, time-varying parameters, non-linearity, exponential family, marked point processes, copulas |
| JEL: | C10 C22 C32 C51 |
| Date: | 2009–03 |
| URL: | http://d.repec.org/n?u=RePEc:hst:ghsdps:gd08-038&r=for |
| By: | Reyes, Celia M; Gonzales, Kathrina G.; Predo, Canesio D.; de Guzman, Rosalina G. |
| Abstract: | <p>Rainfall variability greatly influences corn production. Thus, an accurate forecast is potentially of value to the farmers because it could help them decide whether to grow their corn now or to delay it for the next cropping opportunity. A decision tree analysis was applied in estimating the value of seasonal climate forecast (SCF) information for corn farmers in Isabela.</p> <p>The study aims to estimate the value of SCF to agricultural decisionmakers under climate uncertainty. Historical climatic data of Isabela from 1951 to 2006 from PAGASA and crop management practices of farmers were used in the Decision Support System for Agrotechnology Transfer (DSSAT) to test the potential impact of climate change on corn. The approach is developed for a more accurate SCF and to be able to simulate corn yields for wet and dry seasons under different climatic conditions. While SCF may potentially affect a number of decisions including crop management practices, fertilizer inputs, and variety selection, the focus of the study was on the effect of climate on corn production. Improving SCF will enhance rainfed corn farmers’ decisionmaking capacity to minimize losses brought about by variable climate conditions.</p> |
| Keywords: | seasonal climate forecast (SCF), decision tree analysis, climate uncertainty, Decision Support System for Agrotechnology Transfer (DSSAT) |
| Date: | 2009 |
| URL: | http://d.repec.org/n?u=RePEc:phd:dpaper:dp_2009-05&r=for |
| By: | Stefan Laséen; Lars E.O. Svensson |
| Abstract: | This paper specifies how to do policy simulations with alternative instrument-rate paths in DSGE models such as Ramses, the Riksbank's main model for policy analysis and forecasting. The new element is that these alternative instrument-rate paths are anticipated by the private sector. Such simulations correspond to situations where the Riksbank transparently announces that it plans to implement a particular instrument-rate path and where this announcement is believed by the private sector. Previous methods have instead implemented alternative instrument-rate paths by adding unanticipated shocks to an instrument rule, as in the method of modest interventions by Leeper and Zha (2003). This corresponds to a very different situation where the Riksbank would nontransparently and secretly plan to implement deviations from an announced instrument rule. Such deviations are in practical simulations normally both serially correlated and large, which seems inconsistent with the assumption that they would remain unanticipated by the private sector. Simulations with anticipated instrument-rate paths seem more relevant for the transparent flexible inflation targeting that the Riksbank conducts. We provide an algorithm for the computation of policy simulations with arbitrary restrictions on nominal and real instrument-rate paths for an arbitrary number of periods after which a given policy rule, including targeting rules and explicit, implicit, or forecast-based instrument rules is implemented. When inflation projections are sufficiently sensitive to the real interest-rate path, restrictions on real interest-rate paths provide more intuitive and robust results, whereas restrictions on nominal interest-rate path may provide somewhat counter-intuitive results. |
| JEL: | E52 E58 |
| Date: | 2009–04 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:14902&r=for |