Variation In Forecast Errors Research Articles

주택시장과 거시경제의 관계에 관한 실증분석

The purpose of this paper is to analyze the relationship between the housing market and the macroeconomy using macroeconomic variables, and housing price variables of South Korea from the first quarter of 1991 to the fourth quarter of 2021. In consideration of the macroeconomic transmission mechanism of housing price fluctuations, I built a VAR model with variables such as real GDP, private consumption, housing investment, private credit, policy interest rate, and housing transaction price. In respect of the impulse response, the result of the analysis showed that housing prices rose when real GDP, private consumption, housing investment, and private credit had risen. When housing prices rose, real GDP, prices, private consumption, private credit, and policy interest rates rose. As a result of forecast error variation decomposition, I found that the contribution of changes in housing prices to changes in each variable was not significant, but changes in real GDP and interest rates had a major impact on changes in housing prices. In the historical decomposition I performed, the contribution of policy interest rates to fluctuations in housing prices was also high. In particular, private credit was considerably affected by the rapid increase in housing prices in 2020. Based on the above results, I found that the increase in housing prices had a relatively long-term impact on the credit market but a poor impact on the real economy through increased consumption and housing investment, while I confirmed that the GDP and interest rates had a major effect on changes in housing prices.

A Liquidity Redistribution Effect in Intercorporate Lending: Evidence from Private Firms in Poland

Purpose: We examine the mechanism of intercorporate lending outside the business group, and a reaction of capital expenditures (CAPEX) and capital engagement in other firms to shocks in the provision of such loans. We diagnose the causes and effects of intercorporate lending outside the business group. Design/Methodology/Approach: We use panel data from annual reports (balance sheets and income statements) of 4,600 private Polish companies that provided loans to other firms in the period 2003-2014. We apply the vector autoregression panel model for microeconomic data and analysis of Granger causality, impulse response functions, and forecast error variation decomposition to explore the mechanism of intercorporate loan provision. Findings: Non-financial firms provide loans outside the business group through redistribution of their cash holdings generated from operating activity (cash flow) and long-term bank loans. The provision of loans by non-financial enterprises decreases CAPEX, as a result of the absence of free cash flows that were already used for loan provision. Shareholder loans substitute for capital engagement in other firms. Practical Implications: The findings could assist policymakers to notice that emergency borrowings from other companies are being used to defer defaults and introduce a new credit risk into the business sector. Originality/Value: The redistribution effect of cash holdings and money borrowed from banks provided to unrelated firms outside the business group is dangerous for the stability of the financial system due to the risk that these “indirect borrowers” will default.

Carbon emissions and business cycles

Carbon emissions and real GDP are strongly correlated over the U.S. business cycle. This relationship suggests that macroeconomic shocks inducing cyclical fluctuations in output should also account for the cyclical behavior of emissions and motivates our analysis. We begin by expanding the set of technology shocks in a popular emissions-augmented dynamic stochastic general equilibrium model from the literature, and show that the model generates positive emissions-GDP comovements to each shock through distinct channels. We then estimate the emissions’ response to empirically identified technology shocks using structural vector autoregressions (SVARs). Using the SVARs, we also rank the shocks in terms of explaining the emissions’ forecast error variation. While emissions tend to rise gradually after most shocks, consistent with their theoretical counterparts, the impulse responses are not statistically significant. Unanticipated technology shocks account for less than 10 percent of the variation in emissions. By contrast, anticipated investment technology shocks account for 25 percent of the variation. Government spending and monetary policy shocks account for less than 1 percent. Importantly, close to two thirds of the variation in emissions appears to be due to a structural shock not yet identified in the literature.

Variance Decomposition Networks: Potential Pitfalls and a Simple Solution

Diebold and Yilmaz (2015) recently introduced variance decomposition networks as tools for quantifying and ranking the systemic risk of individual firms. The nature of these networks and their implied rankings depend on the choice decomposition method. The standard choice is the order invariant generalized forecast error variance decomposition of Pesaran and Shin (1998). The shares of the forecast error variation, however, do not add to unity, making difficult to compare risk ratings and risks contributions at two different points in time. As a solution, this paper suggests using the Lanne-Nyberg (2016) decomposition, which shares the order invariance property. To illustrate the differences between both decomposition methods, I analyzed the global financial system during 2001 – 2016. The analysis shows that different decomposition methods yield substantially different systemic risk and vulnerability rankings. This suggests caution is warranted when using rankings and risk contributions for guiding financial regulation and economic policy.

Variance Decomposition Networks: Potential Pitfalls and a Simple Solution

Diebold and Yilmaz (2015) recently introduced variance decomposition networks as tools for quantifying and ranking the systemic risk of individual firms. The nature of these networks and their implied rankings depend on the choice decomposition method. The standard choice is the order invariant generalized forecast error variance decomposition of Pesaran and Shin (1998). The shares of the forecast error variation, however, do not add to unity, making difficult to compare risk ratings and risks contributions at two different points in time. As a solution, this paper suggests using the Lanne-Nyberg (2016) decomposition, which shares the order invariance property. To illustrate the differences between both decomposition methods, I analyzed the global financial system during 2001 – 2016. The analysis shows that different decomposition methods yield substantially different systemic risk and vulnerability rankings. This suggests caution is warranted when using rankings and risk contributions for guiding financial regulation and economic policy.

Variance Decomposition Networks: Potential Pitfalls and Simple Solutions

Diebold and Yilmaz (2015) recently introduced variance decomposition networks as tools for quantifying and ranking the systemic risk of individual firms. The nature of these networks and their implied rankings depend on the choice decomposition method. The standard choice is the order invariant generalized forecast error variance decomposition of Pesaran and Shin (1998). The shares of the forecast error variation, however, do not add to unity, making difficult to compare risk ratings and risks contributions at two different points in time. As a solution, this paper suggests using the Lanne-Nyberg (2016) decomposition, which shares the order invariance property. To illustrate the differences between both decomposition methods, I analyzed the global financial system during 2001 – 2016. The analysis shows that different decomposition methods yield substantially different systemic risk and vulnerability rankings. This suggests caution is warranted when using rankings and risk contributions for guiding financial regulation and economic policy.

Time varying biases and the state of the economy

This paper aims to investigate whether a forecast is optimal, given the information available when it is made. Going beyond the papers that study forecast errors based on the model of Nordhaus (1987), we use a time-varying procedure to forecast revisions and to account for the possibility that the duration of the state may also affect the bias. Three testable hypotheses are presented to help researchers test the optimality of forecasts, with the ultimate aim of determining whether these biases depend on the underlying economic state and whether they are persistent for the duration of the state. Corresponding bias-corrected forecasts can then be made based on these results. The empirical study finds that the one-quarter-ahead official forecast of GDP growth in Taiwan does indeed suffer from state-dependent biases: a persistent under-estimation bias in the relatively good state, and an under-reaction bias that decays with duration in the relatively bad one. Eliminating these biases in the forecast can remove over 44.0% of the variation in forecast errors, and pseudo out-of-sample experiments further support the fact that the resulting bias-corrected forecasts are markedly better than those made by Taiwan’s government or using other competing models.

Determinants and Forecast of Price Level in India: a VAR Framework

The theoretical association of money supply and exchange rates with prices has been empirically established and shown to be dominant in explaining changes in price levels in India. However, post liberalisation, studies have shown price levels to be impacted by several other factors as also, weakened influence of the traditional factors established by theories. This study aims to find the determinants of price level for the period 1994–2008 using a Vector Autoregression model and test the predictive ability of the model. Our results show shorter and smaller impact of change in money supply and nominal effective exchange rate on price levels. Both money supply and nominal effective exchange rates are found to Granger-cause Consumer Price Index. But, impulse response functions show that the impact of shocks from money supply and nominal effective exchange rates on consumer prices peaks after two lags and is short-lived. Forecast error variance decomposition shows that these demand side factors contribute only 6 % of the forecast error variation in Consumer Price Index.

The Prediction of Northern Hemisphere Tropical Cyclone Extended Life Cycles by the ECMWF Ensemble and Deterministic Prediction Systems. Part I: Tropical Cyclone Stage*

Abstract This study has explored the prediction errors of tropical cyclones (TCs) in the European Centre for Medium-Range Weather Forecasts (ECMWF) Ensemble Prediction System (EPS) for the Northern Hemisphere summer period for five recent years. Results for the EPS are contrasted with those for the higher-resolution deterministic forecasts. Various metrics of location and intensity errors are considered and contrasted for verification based on IBTrACS and the numerical weather prediction (NWP) analysis (NWPa). Motivated by the aim of exploring extended TC life cycles, location and intensity measures are introduced based on lower-tropospheric vorticity, which is contrasted with traditional verification metrics. Results show that location errors are almost identical when verified against IBTrACS or the NWPa. However, intensity in the form of the mean sea level pressure (MSLP) minima and 10-m wind speed maxima is significantly underpredicted relative to IBTrACS. Using the NWPa for verification results in much better consistency between the different intensity error metrics and indicates that the lower-tropospheric vorticity provides a good indication of vortex strength, with error results showing similar relationships to those based on MSLP and 10-m wind speeds for the different forecast types. The interannual variation in forecast errors are discussed in relation to changes in the forecast and NWPa system and variations in forecast errors between different ocean basins are discussed in terms of the propagation characteristics of the TCs.

Analysing the Effects of Macroeconomic Shocks on Public Debt in Zimbabwe

This paper analyses the impact of macroeconomic shocks on public debt in Zimbabwe. Understanding the impact of macroeconomic shocks on public debt is necessary as this allows debt managers to focus on a public debt structure which limits the potential of debt from deviating from its long-run steady state. The study applied the Bayesian Vector Auto regression (BVAR) model to simulate the impact of macroeconomic shocks on public debt. The results show that Zimbabwe’s public debt is more vulnerable to interest rate, exchange rate, economic growth and primary balance shocks. Together these shocks account for about 61 per cent of forecast error variation in the debt to GDP ratio. From this analysis, the major policy implication is the need for government to pay particular attention to automatic debt dynamics. There is also a need to maintain the primary balance at manageable levels as well as instituting growth enhancing policies to ensure long-term sustainability of public debt. The need for appropriate selection of the currency composition of public debt is also necessary to mitigate the risk of unexpected increases in public debt from adverse external sector developments.

Estimation of Confidence Intervals of Global Horizontal Irradiance Obtained from a Weather Prediction Model

In this study, we investigated forecast errors of the global horizontal irradiance (GHI) of the numerical weather prediction model developed by the Japan meteorological Agency (JMA) based on the surface-observed data at the JMA stations. We found seasonal variations of forecast errors of the GHI values. From the information of forecast errors, we tried to estimate the confidence intervals of the hourly GHI values. We also compared with results in both for a pinpoint forecast and relatively wide-area forecast. About 68% of the reduction of confidence intervals for the wide-area forecasts compared with the results for the pinpoint forecast is confirmed.

Estimation and forecasting with logarithmic autoregressive conditional duration models: A comparative study with an application

This paper presents a semi-parametric method of parameter estimation for the class of logarithmic ACD (Log-ACD) models using the theory of estimating functions (EF). A number of theoretical results related to the corresponding EF estimators are derived. A simulation study is conducted to compare the performance of the proposed EF estimates with corresponding ML (maximum likelihood) and QML (quasi maximum likelihood) estimates. It is argued that the EF estimates are relatively easier to evaluate and have sampling properties comparable with those of ML and QML methods. Furthermore, the suggested EF estimates can be obtained without any knowledge of the distribution of errors is known. We apply all these suggested methodology for a real financial duration dataset. Our results show that Log-ACD (1,1) fits the data well giving relatively smaller variation in forecast errors than in Linear ACD (1,1) regardless of the method of estimation. In addition, the Diebold–Mariano (DM) and superior predictive ability (SPA) tests have been applied to confirm the performance of the suggested methodology. It is shown that the new method is slightly better than traditional methods in practice in terms of computation; however, there is no significant difference in forecasting ability for all models and methods.

'Some Unpleasant Fiscal Arithmetic': The Role of Monetary and Fiscal Policy in Public Debt Dynamics Since the 1970s

Shocks to monetary and fiscal policy have played a major role in public debt developments in the OECD countries since the mid-1970s. According to the applied VAR approach, these shocks, taken together, explained, on average, about half of the forecast error variation in the debt to GDP ratio, while the share of shocks to GDP growth was close to 30 percent. In contrast, shocks to inflation and the debt ratio itself played in most cases only a minor role. However, the inflation shocks were vital in initiating the public debt problems, as the increase in actual inflation, and particularly the persistence of high inflation expectations in the 1980s, led to a prolonged period of high real interest rates. Learning the implications of greater monetary discipline therefore gave rise to 'some unpleasant fiscal arithmetic' which aggravated debt problems. In most countries fiscal policy aimed at correcting the deterioration in fiscal balances, but the progress was in most cases slow and delayed. It is noticeable that public debt developments have been quite similar in both the United States and the euro area despite differences in fiscal policy and the role of the public sector. Shocks to GDP growth, inflation and monetary policy, which have been more similar in both continents, explain about two thirds of the forecast error variation of the debt to GDP ratio, while shocks to fiscal policy explain about 20 percent.

Wage and Price Dynamics in Portugal

This paper investigates the persistence of aggregate wages and prices in Portugal assuming a model of a unionized economy with imperfect competition. An impulse response analysis is conducted where the structural shocks are identified by taking into account the long-run properties of the model, as well as the cointegrating and weak-exogeneity properties of the system. Real wages and wage inflation emerge as especially persistent following an import price shock, while price inflation is more persistent following an unemployment shock. At the business cycle horizon variation in the forecast errors of wages is attributable mainly to unemployment shocks (about 80 percent), whereas variation in the forecast errors of prices is attributable mainly to import price shocks (about 60 percent) and to unemployment shocks (around 20 percent). Productivity shocks explain somewhat less than 10 percent of the variation in forecast errors of wages and prices.

Public Debt Dynamics in Selected OECD Countries: The Role of Fiscal Stabilisation and Monetary Policy

Shocks to monetary and fiscal policy have played a major role in public debt developments since the mid 1970s. According to the applied VAR approach, together these shocks explained, on average, about half of the forecast error variation in the debt to GDP ratio while the share of shocks to GDP growth was close to 30 percent. Instead, shocks to inflation and the debt ratio itself played in most cases a minor role. However, the inflation shocks were vital in initiating the public debt problems as the increase in actual inflation and particularly the persistence of high inflation expectations in the 1980s led to a prolonged period of high real interest rates. This gave rise to ‘some unpleasant fiscal arithmetic’ which aggravated debt problems. In most countries fiscal policy has aimed at correcting the deterioration of fiscal balances, but the progress has in most cases been slow and delayed. Nevertheless, all individual country VARs are stable in the period under consideration. Finally, contrary to general beliefs, in the global financial markets of present day inflation makes debt problems worse through its adverse impact on interest rates.

Modeling the accuracy of certain regional predictions

A model is hypothesized specifying forecast error as a function of specific ‘year effects’, particular dates of forecast, and ‘time span effects’, length of projection horizon. Model estimation methodology is presented and empirical application is made to inaccuracies arising in the RAS procedure for projecting bivariate region/industry employment arrays for two sets of regional data. In both applications, the model is found to explain a large proportion of variation in forecast errors, and estimation results permit isolation of pure time effects on deterioration of model accuracy and quantification of relative ‘year effects’.

An Analysis of Western North Pacific Tropical Cyclone Forecast Errors

Western North Pacific tropical cyclone position forecast errors for 10 years (1966–75) are statistically analyzed. Variations of errors versus a number of parameters are examined. It is shown that a small number of readily available parameters, such as location, maximum wind and components of motion, can, with reasonable effectiveness, classify a tropical cyclone forecast as representing a group with either markedly above or below average errors. The annual variations of forecast errors are also discussed and an attempt is made to explain those variations.