In‐Sample and Out‐of‐Sample Prediction of stock Market Bubbles: Cross‐Sectional Evidence

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in AbstractWe evaluate the informational content of ex post and ex ante predictors of periods of excess stock (market) valuation. For a cross section comprising 10 OECD economies and a time span of at most 40 years alternative binary chronologies of price bubble periods are determined. Using these chronologies as dependent processes and a set of macroeconomic and financial variables as explanatory variables, logit regressions are carried out. With model estimates at hand, both in-sample and out-of-sample forecasts are made. Overall, the degree of ex ante predictability is limited if an analyst targets the detection of particular turning points of market valuation. The set of 13 potential predictors is classified in measures of macroeconomic or monetary performance, stock market characteristics, and descriptors of capital valuation. The latter turn out to have strongest in-sample and out-of-sample explanatory content for the emergence of price bubbles. In particular, the price to book ratio is fruitful to improve the ex-ante signalling of stock price bubbles.Keywords: Stock market bubbles; out-of-sample forecasting; financial ratios; OECD countries.JEL classification: G01; G17; E27. ¶ We thank Christian Dreger and Hans-Eggert Reimers for their helpful suggestions. Moreover, we thank Martin Everts for the provision of his Matlab code designed to implement the quarterly Bry-Boschan procedure for the identification of stock price bubbles and Matthias Hartmann and Wieland Hoffmann for computational support and data handling.

Uncertainty of Macroeconomic Forecasters and the Prediction of Stock Market Bubbles

“…To compare the forecast accuracy of alternative models we use a modified Diebold-Mariano statistic developed by Harvey et al (1997 Regarding the estimated impacts of macroeconomic, monetary, and stock market variables the results are generally in line with Herwartz and Kholodilin (2014). Both the variance of the forecasts of the real GDP growth rate and the corresponding positive semivariance are statistically significant and have negative signs.…”

“…Two chronologies are considered: a more liberal (φ = 1) indicating boom or bubble events (for ≈16% of all sample observations), and a conservative chronology (φ = 1.5) indicating bubble periods (for ≈7.5% of sample observations). Similar to Herwartz and Kholodilin (2014) we rely on three groups of predictors describing i) macroeconomic situation (real GDP growth, current account balance-to-GDP ratio), ii) credit market conditions (real money market interest rate, term spread), and iii) stock market variables (returns, volatility) (see Table 1 for variable definitions and sources). The considered cross section consists of France, Germany, Italy, Japan, UK, and the USA, and the data cover the period from 1989 Q1 to 2014 Q2.…”

“…For model specification we build upon earlier work in Herwartz and Kholodilin (2014), and evaluate the marginal contribution of macroeconomic uncertainty to the forecast accuracy of panel logit regressions.…”

Early Warning Indicator Model of Financial Developments Using an Ordered Logit

“…This is also done by Gerdesmeier et al (2010, 2011), Herwartz & Kholodilin (2011 or Dreger & Kholodilin (2011). As for the financial variables, we consider historical series of the short-term (three-month money market) and long-term (ten-year government bond yield) interest rates and their spreads.…”

“…An overview is also given by Gerdesmeier, Reimers and Roffia (2011). Early warning model approaches are applied to analyse bubbles on stock markets as in Herwartz and Kholodilin (2011), whereas Angello and Schuknecht (2011) and Dreger and Kholodilin (2011) concentrate on house markets. Both markets are investigated by Borio and Lowe (2002) or Helbling and Terrones (2003) or as a composite asset constructed from these markets by Alessi and Detken (2009), Borio et al or Gerdesmeier et al (2010, 2011.…”