Forecasting banking crises with dynamic panel probit models

Banking crises are rare events, but when they occur, their consequences are often dramatic. The aim of this paper is to contribute to the toolkit of early warning models that is available to policy makers by exploring the dynamics and exuberances embedded in a panel dataset that covers 22 European countries over four decades (from 1970Q1 to 2012Q4). The in- and out-of-sample forecast performances of several (dynamic) probit models are evaluated, with the objective of developing common vulnerability indicators with early warning properties. The results obtained show that adding dynamic components and exuberance indicators to the models improves the performances of early warning models significantly.     

Systemic event prediction by an aggregate early warning system: An application to the Czech Republic

This work develops an early warning framework for assessing systemic risks and predicting systemic events over a short horizon of six quarters and a long horizon of 12 quarters on a panel of 14 countries, both advanced and developing. First, we build a financial stress index to identify the starting dates of systemic financial crises for each country in the panel. Second, early warning indicators for the assessment and prediction of systemic risk are selected in a two-step approach; we find relevant prediction horizons for each indicator by a univariate logit model followed by the application of Bayesian model averaging to identify the most useful indicators. Finally, we observe the performance of the constructed EWS over both horizons on the Czech data and find that the model over the long horizon outperforms the EWS over the short horizon. For both horizons, out-of-sample probability estimates do not deviate substantially from their in-sample estimates, indicating a good out-of-sample performance for the Czech Republic.     

Liquidity indicators,early warning signals in banks,and financial crises

This research compares the performance of three liquidity indicators, namely liquidity ratio (LiqR), liquidity creation (LiqC) and net stable funding difference (NSFD), for sending early warning signals for distressed banks. Recent evidence has shown that LiqR appears incapable of measuring the liquidity condition of banks. However, LiqC and NSFD have not yet been fully examined. Thus, which indicator is more useful in an early warning model becomes an interesting issue. We classify distressed banks as banks that have experienced a bank run, bailout, or failure. Sample data are collected from the United States and the European Union from before and after the financial crisis. We then estimate model predictive value using the sample before the crisis to predict liquidity shortages. Evidence shows that the academic (LiqC) and officially recommended indicators (NSFD) outperform LiqR as early warning signals. Furthermore, LiqC performs best when banks actively engage in income diversification but not fund diversification. Therefore, a well income-diversified bank with high LiqC tends to have high distress probability in the next period.     

Behavioural heterogeneity and shift-contagion: Evidence from the Asian crisis

In this paper, we propose an empirical model based on the heterogeneous agents literature. Price changes are induced by fundamental, technical, and international factors. The model is estimated for Hong Kong and Thailand surrounding the Asian crisis. We find that the three sources are relevant and that their relative price impact fluctuates conditional on price impact in the previous period. Results imply that the crisis is triggered in Thailand due to an increased focus on the fundamental price, followed by an increase in chartism and finally aggravated by a focus on foreign developments. Furthermore, the crisis deepens in Hong Kong because of increased attention for foreign markets.     

Evaluating early warning indicators of banking crises: Satisfying policy requirements

Ideally, early warning indicators (EWI) of banking crises should be evaluated on the basis of their performance relative to the macroprudential policy maker’s decision problem. We translate several practical aspects of this problem — such as difficulties in assessing the costs and benefits of various policy measures, as well as requirements for the timing and stability of EWIs — into statistical evaluation criteria. Applying the criteria to a set of potential EWIs, we find that the credit-to-GDP gap and a new indicator, the debt service ratio (DSR), consistently outperform other measures. The credit-to-GDP gap is the best indicator at longer horizons, whereas the DSR dominates at shorter horizons.     

Real estate risk measurement and early warning based on PSO-SVM

The measurement and early warning of real estate risk are important to prevent and defuse major financial risks, and they form a basis for high-quality development. This paper assessed the internal and external environments of the real estate market; constructed a real estate risk indicator system from the aspects of market level, real estate enterprises, policy factors and financial institutions; and implemented a PSO-SVM model to measure and warn of real estate risk. Empirical studies were conducted. The results show the following: (1) the synthetic real estate risk index well depicts the cyclical fluctuation of real estate risk in Beijing; (2) the warning model based on the PSO-SVM method exhibits better performance and higher warning accuracy than other models do.     

Penalized maximum likelihood estimation of logit-based early warning systems

Panel logit models have proved to be simple and effective tools to build early warning systems (ews) for financial crises. But because crises are rare events, the estimation of ews does not usually account for country-specific fixed effects, so as to avoid losing all the information relative to countries that never face a crisis. I propose using a penalized maximum likelihood estimator for fixed-effects logit-based ews where all the observations are retained. I show that including country effects, while preserving the entire sample, improves the predictive performance of ews, both in simulation and out of sample, with respect to the pooled, random-effects and standard fixed-effects models.     

Credit-based early warning indicators of banking crises in emerging markets

This article explores the role of credit-based variables as early warning indicators (EWIs) of banking crises in the context of emerging economies. We collect data on bank and total credit to the private sector in emerging markets and evaluate the signalling performance by using the area under the receiver operating characteristics (ROC) curve (AUC). Our results show that nominal credit growth and the change in the credit-to-GDP ratio have the best signalling properties and significantly outperform the credit-to-GDP gap in almost all specifications for policy-relevant horizons. These findings are in stark contrast with the results on advanced economies, where the credit-to-GDP gap is the single best performing EWI. Our results emphasize the importance of caution when applying statistical methods calibrated for advanced markets to emerging economies.     

Structural vulnerability and resilience to currency crisis: Foreign currency debt versus export

Is there any factor that is not analyzed in the literature but is important for preventing currency crises? I argue that exports are an important factor to prevent currency crises that has not been frequently analyzed in the existing theoretical literature. Using the third generation model of currency crises, I derive a simple and intuitive formula that captures an economy’s structural vulnerability characterized by the elasticity of exports and repayments for foreign currency denominated debt. I graphically show that the possibility of currency crisis equilibrium depends on this structural vulnerability and also analyze how this vulnerability impacts the effectiveness of monetary policy response.     

Systemic financial risk early warning of financial market in China using Attention-LSTM model

We propose an Attention-LSTM neural network model to study the systemic risk early warning of China. Based on text mining, the network public opinion index is constructed and used as a training set to be incorporated into the early warning model to test the early warning effect. The results show that: (i) the network public opinion is the non-linear Granger causality of systemic risk. (ii) The Attention-LSTM neural network has strong generalization ability. Early warning effects have been significantly improved. (iii) Compared with the BP neural network model, the SVR model and the ARIMA model, the LSTM neural network early warning model has a higher accuracy rate, and its average prediction accuracy for systemic risk indicators has been improved over short, medium and long terms. When the attention mechanism is included in the LSTM, the Attention-LSTM neural network model is even more accurate in all the cases.     

基于GPRS/CDMA通信的山洪灾害预警系统的设计

文章介绍了利用CDMA/GPRS网络,借助CDMA/GPRS通讯产品,完成各级山洪预警系统中的监测采集站网的汛情信息传输、监测中心的监测信息传送以及预警广播信息的传送,为山洪灾害的及时监测、预警预报,以及决策会商和指挥调度工作提供了畅通、快速、可靠稳定的通信信道。     

温度实时监测及预警系统在煤矿机电设备中的应用研究

随着煤矿生产的机械化及电子化程度越来越高,由机电设备引发的煤矿火灾事故接连频繁的发生,文章通过对火灾产生原因进行分析,利用光纤光栅进行温度实时监测及预警的安全系统,为煤矿机电设备提供全方位、及时、准确的监控,同时,为工作人员的安全管理提供可靠性辅助决策,使煤矿机电设备事故的发生得到一定的预防及控制,有助于煤矿生产顺利安全的进行。     

Forecasting with approximate dynamic factor models: The role of non-pervasive shocks

This paper studies the role of non-pervasive shocks when forecasting with factor models. To this end, we first introduce a new model that incorporates the effects of non-pervasive shocks, an Approximate Dynamic Factor Model with a sparse model for the idiosyncratic component. Then, we test the forecasting performance of this model both in simulations, and on a large panel of US quarterly data. We find that, when the goal is to forecast a disaggregated variable, which is usually affected by regional or sectorial shocks, it is useful to capture the dynamics generated by non-pervasive shocks; however, when the goal is to forecast an aggregate variable, which responds primarily to macroeconomic, i.e. pervasive, shocks, accounting for non-pervasive shocks is not useful.     

The international transmission channels of US supply and demand shocks: Evidence from a non-stationary dynamic factor model for the G7 countries

We employ a multi-country non-stationary dynamic factor model to assess spillover effects and transmission channels of US supply and demand shocks on a variety of macroeconomic variables in individual non-US G7 countries. We find that trade, financial and confidence channels all play a significant role in the international transmission of US shocks. However, the results point to substantial heterogeneities of shock transmission across the individual G7 economies. In particular, we find negative transmission effects for Italy and Japan as the only two G7 countries not well integrated into global value chains. Moreover, the exchange rate responses of Germany, France and Italy turn out to be far less pronounced in comparison to the other G7 economies which we relate to their membership of the euro area and their coordinated monetary policies prior to the establishment of the euro. Whereas we document a close comovement of stock market dynamics across the G7 countries, we find credit and real estate markets to be less synchronized. We do not find the effects and transmission channels to be fundamentally affected by the post-2008 economic environment.     

30 years of cointegration and dynamic factor models forecasting and its future with big data: Editorial

The seed of this special section was the workshop celebrated at FUNCAS in Madrid in February 2019 “30 Years of Cointegration and Dynamic Factor Models Forecasting and its Future with Big Data”. In this editorial, we describe the main contributions of the 13 papers published within the special section towards forecasting in the context of non- stationary Big Data using cointegration or Dynamic Factor Models.     

Budgeting for research and development: a dynamic financial simulation approach

The research and development (R&D) budgeting decision is crucial for at least two reasons: if too much is spent, short-term financial stability is at risk, while, if the budget is too small, long-term competitiveness is threatened. Nevertheless, many enterprises simply extrapolate the past without further reflection.This paper presents a computer-based dynamic stochastic simulation model that allows one to assess the impact of alternative R&D budgeting policies on corporate development. The core decisions to be evaluated concern timing and funding of investments in R&D. Our approach substantially expands earlier work by Brockhoff (R&D Manage. 19 (1989) 265). In particular, it distinguishes between product and process innovation, considers market dynamics related to technical progress via a modifiable S-curve, integrates marketing, and takes into account essential financial aspects. As a result, our model is closer to reality than previous ones. A sample application with real company data illustrates its potential usage.