|
on Forecasting |
| By: | Kerstin Bernoth; Andreas Pick |
| Abstract: | This paper considers the issue of forecasting financial fragility of banks and insurances using a panel data set of performance indicators, namely distance-to- default, taking unobserved common factors into account. We show that common factors are important in the performance of banks and insurances, analyze the influences of a number of observable factors on banking and insurance performance, and evaluate the forecasts from our model. We find that taking unobserved common factors into account reduces the the root mean square forecasts error of firm specific forecasts by up to 11% and of system forecasts by up to 29% relative to a model based only on observed variables. Estimates of the factor loadings suggest that the correlation of financial institutions has been relatively stable over the forecast period. |
| Keywords: | Financial stability, financial linkages, banking, insurances, unobserved common factors, forecasting |
| JEL: | C53 G21 G22 |
| Date: | 2009 |
| URL: | http://d.repec.org/n?u=RePEc:diw:diwwpp:dp882&r=for |
| By: | Söderberg, Jonas (Centre for Labour Market Policy Research (CAFO)) |
| Abstract: | This paper evaluates 14 macroeconomic variables’ ability to forecast changes in monthly liquidity on the Scandinavian order-driven stock exchanges. Every macroeconomic variable is evaluated both out-of-sample and in-sample and against three different benchmark models of market variables and asymmetries concerning up and down markets. Policy rate on Copenhagen, broad money growth on Oslo, and short-term interest rate and flows from mutual funds on Stockholm significantly improve the out-of-sample forecasts of liquidity at these exchanges. However, most proposed macroeconomic variables can be rejected as forecasters of liquidity on the Scandinavian stock exchanges. There are many variables that predict in-sample liquidity that do not forecast out-of-sample. This stresses the importance of conducting out-of-sample tests when examining whether macroeconomic variables predict liquidity. In addition, this is the first paper confirming that stock market liquidity can be forecast out-of-sample. |
| Keywords: | Liquidity; Scandinavian stock markets; Forecasting; Out-of-sample tests. |
| JEL: | G12 G18 |
| Date: | 2008–12–01 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:vxcafo:2009_010&r=for |
| By: | Michele Modugno (European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.); Kleopatra Nikolaou (European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.) |
| Abstract: | This paper investigates whether information from foreign yield curves helps forecast domestic yield curves out-of-sample. A nested methodology to forecast yield curves in domestic and international settings is applied on three major countries (the US, Germany and the UK). This novel methodology is based on dynamic factor models, the EM algorithm and the Kalman lter. The domestic model is compared vis-á-vis an international one, where information from foreign yield curves is allowed to enrich the information set of the domestic yield curve. The results have interesting and original implications. They reveal clear international dependency patterns, strong enough to improve forecasts of Germany and to a lesser extent UK. The US yield curve exhibits a more independent behaviour. In this way, the paper also generalizes anecdotal evidence on international interest rate linkages to the whole yield curve. JEL Classification: F31. |
| Keywords: | Yield curve forecast, Dynamic factor model, EM algorithm, International linkages. |
| Date: | 2009–04 |
| URL: | http://d.repec.org/n?u=RePEc:ecb:ecbwps:200901044&r=for |
| By: | Mukherjee, Deepraj; Kemme, David |
| Abstract: | Countries in which inflation targeting has been adopted require high quality inflation forecasts. The Polish National Bank adopted a variant of implicit inflation targeting and therefore the ability to forecast inflation is critically important to policy makers. Since the domestic price formation process is still evolving, medium term inflation forecasting is often difficult. Using quarterly data from 1995-2007, we estimate and evaluate three types of models for inflation forecasting: (1) output gap models, (2) models involving money, and (3) models which bring the foreign sector into the price formation process. We find that openness is significant in the price formation process and inflation targeting is associated with lower inflation. Traditional measures of forecast accuracy indicate that the simple price gap version of the P* model and the money demand model perform best of this group for medium term forecasting. |
| Keywords: | Monetary policy; inflation; forecasting; models. |
| JEL: | E0 E31 E52 E37 |
| Date: | 2008–07–12 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:14952&r=for |
| By: | Chauvet, Marcelle; Senyuz, Zeynep |
| Abstract: | This paper proposes an econometric model of the joint dynamic relationship between the yield curve and the economy to predict business cycles. We examine the predictive value of the yield curve to forecast both future economic growth as well as the beginning and end of economic recessions at the monthly frequency. The proposed multivariate dynamic factor model takes into account not only the popular term spread but also information extracted from the entire yield curve. The nonlinear model is used to investigate the interrelationship between the phases of the bond market and of the business cycle. The results indicate a strong interrelation between these two sectors. Although the popular term spread has a reasonable forecasting performance, the proposed factor model of the yield curve exhibits substantial incremental predictive value. This result holds in-sample and out-of-sample, using revised or real time unrevised data. |
| Keywords: | Forecasting; Business Cycles; Yield Curve; Dynamic Factor Models; Markov Switching. |
| JEL: | C32 E32 E44 |
| Date: | 2008–12 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:15076&r=for |
| By: | Lucia Alessi (Directorate General Statistics, European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.); Carsten Detken (Directorate General Research, European Central Bank, Kaiserstrasse 29, D-60311 Frankfurt am Main, Germany.) |
| Abstract: | We test the performance of a host of real and financial variables as early warning indicators for costly aggregate asset price boom/bust cycles, using data for 18 OECD countries between 1970 and 2007. A signalling approach is used to predict asset price booms that have relatively serious real economy consequences. We use a loss function to rank the tested indicators given policy makers’ relative preferences with respect to missed crises and false alarms. The paper analyzes the suitability of various indicators as well as the relative performance of financial versus real, global versus domestic and money versus credit based liquidity indicators. We find that global measures of liquidity are among the best performing indicators and display forecasting records,, which provide useful information for policy makers interested in timely reactions to growing financial imbalances, as long as aversion against type I and type II errors is not too unbalanced. Furthermore, we explore out-of-sample whether the most recent wave of asset price booms (2005-2007) would be predicted to be followed by a serious economic downturn. JEL Classification: E37, E44, E51. |
| Keywords: | Early Warning Indicators, Signalling Approach, Leaning Against the Wind, Asset Price Booms and Busts, Global Liquidity. |
| Date: | 2009–03 |
| URL: | http://d.repec.org/n?u=RePEc:ecb:ecbwps:200901039&r=for |
| By: | Peroni, Chiara |
| Abstract: | This paper studies determinants of risk premia using a non-parametric term-structure model of the corporate spread. The model, which measures the extra return of defaultable corporate bonds on their government counterparts, involves the rate of inflation, a key macroeconomic variable that is found to explain the spread non-linearly. This study shows that non-linear methods are useful to investigate features of credit risk and that they give better results than their linear counterparts, enabling testing of affine term-structure specifications. The paper also shows how the non-linear model can be used to forecast the future course of the spread. |
| Keywords: | risk premium; corporate spread; default; additive models; non-parametric estimation. |
| JEL: | G12 C14 E44 |
| Date: | 2008–02–11 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:15010&r=for |