|
on Forecasting |
| By: | Snowberg, Erik (California Institute of Technology); Wolfers, Justin (Wharton School, University of Pennsylvania); Zitzewitz, Eric (Dartmouth College) |
| Abstract: | Prediction markets – markets used to forecast future events – have been used to accurately forecast the outcome of political contests, sporting events, and, occasionally, economic outcomes. This chapter summarizes the latest research on prediction markets in order to further their utilization by economic forecasters. We show that prediction markets have a number of attractive features: they quickly incorporate new information, are largely efficient, and impervious to manipulation. Moreover, markets generally exhibit lower statistical errors than professional forecasters and polls. Finally, we show how markets can be used to both uncover the economic model behind forecasts, as well as test existing economic models. |
| Keywords: | prediction markets, forecasting |
| JEL: | C5 G14 |
| Date: | 2012–07 |
| URL: | http://d.repec.org/n?u=RePEc:iza:izadps:dp6720&r=for |
| By: | Pablo Pincheira; Carlos Medel |
| Abstract: | The use of different time-series models to generate forecasts is fairly usual in the forecasting literature in general, and in the inflation forecast literature in particular. When the predicted variable is stationary, the use of processes with unit roots may seem counterintuitive. Nevertheless, in this paper we demonstrate that forecasting a stationary variable with driftless unit-root-based forecasts generates bounded Mean Squared Prediction Errors errors at every single horizon. We also show via simulations that persistent stationary processes may be better predicted by unit-root-based forecasts than by forecasts coming from a model that is correctly specified but that is subject to a higher degree of parameter uncertainty. Finally we provide an empirical illustration in the context of CPI inflation forecasts for three industrialized countries. |
| Date: | 2012–07 |
| URL: | http://d.repec.org/n?u=RePEc:chb:bcchwp:669&r=for |
| By: | Norbert Schanne |
| Abstract: | The development of employment and unemployment in regional labour markets is known to spatially interdependent. Global Vector-Autoregressive (GVAR) models generate a link between the local and the surrounding labour markets and thus might be useful when analysing and forecasting employment and unemployment even if they are non-stationary or co-trending. Furthermore, GVARs have the advantage to allow for both strong cross-sectional dependence on ``leader regions' and weak cross-sectional, spatial dependence. For the recent and further development of labour markets the economic situation (described e.g. by business-cycle indicators), politics and environmental impacts (e.g. climate) may be relevant. Information on these impacts can be integrated in addition to the joint development of employment and unemployment and the spatial link in a way that allows on the one hand to carry out economic plausibility checks easily and on the other hand to directly receive measures regarding the statistical properties and the precision of the forecasts. Then, the forecasting accuracy is demonstrated for German regional labour-market data in simulated forecasts at different horizons and for several periods. Business-cycle indicators seem to have no information regarding labour-market prediction, climate indicators little. In contrast, including information about labour-market policies and vacancies, and accounting for the lagged and contemporaneous spatial dependence can improve the forecasts relative to a simple bivariate model. |
| Date: | 2011–09 |
| URL: | http://d.repec.org/n?u=RePEc:wiw:wiwrsa:ersa10p1044&r=for |
| By: | D'Elia, Enrico |
| Abstract: | In general, rational economic agents trade off the cost of waiting for the statistical agencies disseminate the final results of the relevant surveys before making a decision, on the one hand, and of making use of some model based predictions. Thus, from the viewpoint of agents, predictions and preliminary results from surveys often compete against each other. Comparing the loss attached to predictions, on the one hand, and to possible preliminary estimate from incomplete samples, on the other, provides a broad guidance in deciding if and when statistical agencies should release preliminary and final estimates. In this paper, the case of the dissemination of figures on quarterly GDP in the Euro Area is examined. The main conclusion is that the so called “flash estimates” actually provide valuable information to the users, while intermediate releases, published before three months from the end of the reference quarter can be substituted by model based estimation without any loss of accuracy. |
| Keywords: | Accuracy; Data Dissemination; Forecast; Nowcast; Preliminary Estimates; Timeliness |
| JEL: | C44 C82 |
| Date: | 2012–07 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:40264&r=for |
| By: | Simon Cornée (University of Rennes 1 - CREM, UMR CNRS 6211) |
| Abstract: | Based on a unique hand-collected database of 389 loans obtained from a French social bank dealing with small businesses, this paper compares two predictive models of future default events: the first relies on soft information (SI model), the second on hard information (HI model). The results indicate that the SI model outperforms the HI model in terms of forecast quality and goodness of fit. In so doing, this paper provides further empirical evidence that, when they serve small businesses, small or decentralized banks have a greater ability to collect and act on soft information. This empirical conclusion conveys practical implication for social banks’ internal credit rating procedures, especially in their calibration of capital requirements. |
| Keywords: | Credit Rating, Debt Default, Small Business Lending, Relationship Lending, Social Banking |
| JEL: | G21 M21 |
| Date: | 2012–06 |
| URL: | http://d.repec.org/n?u=RePEc:tut:cremwp:201226&r=for |
| By: | DNB |
| Abstract: | This Occasional Study presents DELFI, a new macroeconomic model of the Dutch economy for forecasting and policy analysis. Macroeconomic modelling at de Nederlandsche Bank started some 25 years ago, when Martin Fase and his team built MORKMON, which was quite novel at the time due to the inclusion of a monetary sector. For many years, this model was used fruitfully as an instrument for forecasting, scenario analysis and policy simulation, and its acronym survived during all those years. But the times have changed and so has the economic environment. Already when MORKMON was introduced in 1984, the then president of de Nederlandsche Bank, Wim Duisenberg, noted how our understanding of the economy is never perfect. Indeed, changes in the economic environment and new insights led to various minor and major adjustments to the structure of MORKMON, up to a point where it was decided to build a completely new model. DELFI is the result of a collective effort by researchers and statisticians at the Economics & Research Division of de Nederlandsche Bank. In my view, the team has created a worthy successor of MORKMON. And while history learns that a model is never perfect or finished, I am confident ELFI is in a position to take over the prominent position that MORKMON had so many years within the set of analytical instruments of de Nederlandsche Bank. |
| Date: | 2011–02 |
| URL: | http://d.repec.org/n?u=RePEc:dnb:dnbocs:901&r=for |
| By: | Anna Norin |
| Abstract: | Business cycles are usually defined at a national level. The implicit assumption being that it affects all regions similarly. This is combined with a lack of timely information on regional economic development as annual values of the gross regional product (GRP) are often published with up to two years lag. The present paper evaluates a method of obtaining values of the GRP as soon as monthly and quarterly business cycle indicators become available. Building on earlier work on using bridge equations to obtaining quarterly values of GDP growth, a method is proposed were annual GRP growth is estimated using a large number of business cycle indicators. The procedure is applied to data for the Northern regions of Sweden. With the present method it is possible to continuously refine GRP growth values throughout the year. By utilizing the information content in available business cycle indicators, a nowcast of the GRP is obtained as opposed to a pure forecast based solely on past information. Nowcasts will then provide valuable information on how current highs or lows are affecting different regions. |
| Date: | 2011–09 |
| URL: | http://d.repec.org/n?u=RePEc:wiw:wiwrsa:ersa10p768&r=for |
| By: | Alexander Guarín; Andrés González; Daphné Skandalis; Daniela Sánchez |
| Abstract: | In this paper, we propose an alternative methodology to determine the existence of credit booms, which is a complex and crucial issue for policymakers. In particular, we exploit the Mendoza and Terrones (2008)’s idea that macroeconomic aggregates other than the credit growth rate contain valuable information to predict credit boom episodes. Our econometric method is used to estimate and predict the probability of being in a credit boom. We run empirical exercises on quarterly data for six Latin American countries between 1996 and 2011. In order to capture simultaneously model and parameter uncertainty, we implement the Bayesian model averaging method. As we employ panel data, the estimates may be used to predict booms of countries which are not considered in the estimation. Overall, our findings show that macroeconomic variables contain valuable information to predict credit booms. In fact, with our method the probability of detecting a credit boom is 80%, while the probability of not having false alarms is greater than 92%. |
| Date: | 2012–07–22 |
| URL: | http://d.repec.org/n?u=RePEc:col:000094:009826&r=for |