Publications
 
Downloadable Version

   
Theses
  • Diploma: Vergleich verschiedener Schätzfunktionen für ein Vorhersagerisiko im Regressionsmodell als Grundlage für empirische Modellwahlverfahren. Humboldt-Universität, Berlin (1980).
  • PhD: Kriterien zur Wahl von Regressionsmodellen. Humboldt-Universität, Berlin (1982). 
  • Habilitation: Modellwahl und verwandte Methoden in der Regressionsanalyse. Humboldt-Universität, Berlin (2003).
   
Refereed publications
  • Bunke, O. and Droge, B. (1984). Bootstrap and cross-validation estimates of the prediction error for linear regression models. Annals of Statistics 12, 1400-1424.
  • Bunke, O. and Droge, B. (1984). Estimators of the mean squared error of prediction in linear regression. Technometrics 26, 145-155.
  • Bunke, H. and Droge, B. (1985). A stepwise procedure for the selection of nonlinear regression models. Statistics 16, 35-45.
  • Droge, B. und Polzehl, J. (1986). "HU-PP Statistik" als multivalent nutzbare Software. Wiss. Zeitschr. TU Dresden 35, 136-138.
  • Droge, B. (1987). A note on estimating the MSEP in nonlinear regression. Statistics 18, 499-520.
  • Droge, B. (1992). On a computer program for the selection of variables and models in regression analysis. In: Model Oriented Data-Analysis (Eds. V. Fedorov, W.G. Müller and I.N. Vuchkov), Physica, Heidelberg, 181-192.
  • Droge, B. (1993). On fnite-sample properties of adaptive least squares regression estimates. Statistics 24, 181-203.
  • Droge, B. and Georg, T. (1995). On selecting the smoothing parameter of least squares regression estimates using the minimax regret approach. Statistics & Decisions 13, 1-20.
  • Droge, B. (1995). Some simulation results on cross-validation and competitors for model choice. In: MODA4 - Advances in Model Oriented Data-Analysis (Eds. C.P. Kitsos and W.G. Müller), Physica, Heidelberg, 213-222.
  • Droge, B. (1996). Some comments on cross-validation. In: Statistical Theory and Computational Aspects of Smoothing (Eds. W. Härdle and M.G. Schimek), Physica, Heidelberg, 178-199.
  • Bunke, O., Droge, B. and Polzehl, J. (1998). Splus tools for model selection in nonlinear regression. Computational Statistics 13, 257-281.
  • Droge, B. (1998). Minimax regret analysis of orthogonal series regression estimation: Selection versus shrinkage. Biometrika 85, 631-643.
  • Droge, B. (1999). Asymptotic optimality of full cross-validation for selecting linear regression models. Statistics & Probability Letters 44, 351-357.
  • Bunke, O., Droge, B. and Polzehl, J. (1999). Model selection, transformations and variance estimation in nonlinear regression. Statistics 33, 197-240.
  • Bunke, O., Droge, B. und Schwalbach, J. (2002). Die relative Bedeutung von Firmen- und Industriezweigeffekten für den Unternehmenserfolg. Zeitschrift für Betriebswirtschaft, 72, H. 3, 275-294.
  • Droge, B. (2006). Minimax regret comparison of hard and soft thresholding for estimating a bounded normal mean. Statistics & Probability Letters 76, 83-92.
  • Droge, B. (2006). Asymptotic properties of model selection procedures in linear regression. Statistics 40, 1-38.
  • Karaman Örsal, D. and Droge, B. (2011). Corrigendum to "Likelihood-Based Cointegration Tests in Heterogeneous Panels". Econometrics Journal 14, 121-125
  • Karaman Örsal, D. and Droge, B. (2014). Panel Cointegration Testing in the
    Presence of a Time Trend. Computational Statistics and Data Analysis 76, 377-390.
   
Other publications
  • Bunke, O. and Droge, B. (1982). Some comparisons of criteria for selecting linear regression models. Preprint Nr. 27, Sektion Mathematik, Humboldt-Universität, Berlin. (A revised version appeared under the title "Estimators of the mean squared error of prediction in linear regression" in Technometrics 26 (1984), 145-155.)
  • Bunke, O. and Droge, B. (1982). Bootstrap and cross-validation criteria for selecting linear regression models. Preprint Nr. 33, Sektion Mathematik, Humboldt-Universität, Berlin. (A revised version appeared under the title "Bootstrap and cross-validation estimates of the prediction error for linear regression models" in Annals of Statistics 12 (1984), 1400-1424.)
  • Droge, B. (1984). A note on estimating the MSEP in nonlinear regression. Preprint Nr. 66, Sektion Mathematik, Humboldt-Universität, Berlin. (Appeared in Statistics 18 (1987), 499-520.)
  • Droge, B. (1986). Ein Programm zur automatisierten Wahl unter nichtlinearen Regressionsmodellen mit einer Einflußgröße. In: "HU-PP Statistik" - Eine verfahrensorientierte Beschreibung einer Sammlung multivalent nutzbarer Statistik-Software (Eds. B. Droge und J. Polzehl), Seminarbericht Nr. 83, Sektion Mathematik, Humboldt-Universität, Berlin, 35-72.
  • Droge, B. (1986). On estimating the prediction error of nonlinear regression models. In: "Nichtlineare Regression" (Eds. O. Bunke and E.Jolivet), Seminarbericht Nr. 89, Sektion Mathematik, Humboldt-Universität, Berlin, 130-152.
  • Droge, B. (1989). Estimating the prediction risk of regression estimators and model selection. Proc. 6th European Young Statistician Meeting, (Eds. M. Hala and M. Maly), Prague, August 1989, 73-80.
  • Droge, B. (1992). On finite-sample properties of some nonparametric regression estimates. Preprint Nr. 92-1, Fachbereich Mathematik, Humboldt-Universität, Berlin. (A revised version appeared under the title "On finite-sample properties of adaptive least squares regression estimates" in Statistics 24 (1993), 181-203.)
  • Bunke, O., Droge, B. and Polzehl, J. (1993). Model selection and variable transformations in nonlinear regression. CORE Discussion Paper No. 9327, C.O.R.E., UCL, Belgium.
  • Droge, B. and Georg, T. (1993). On selecting the smoothing parameter of least squares regression estimates using the minimax regret approach. Preprint Nr. 93-10, Fachbereich Mathematik, Humboldt-Universität, Berlin. (Appeared in Statistics & Decisions 13 (1995), 1-20.)
  • Droge, B. (1994). Some comments on cross-validation. Discussion Paper No. 94-7, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (Appeared in: Statistical Theory and Computational Aspects of Smoothing, Eds. W. Härdle and M.G. Schimek, Physica, Heidelberg, 178-199, 1996.)
  • Droge, B. (1994). Some simulation results on cross-validation and competitors for model choice. Discussion Paper No. 94-30, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (Appeared in: MODA4 - Advances in Model Oriented Data-Analysis, Eds. C.P. Kitsos and W.G. Müller, Physica, Heidelberg, 213-222, 1995.)
  • Bunke, O., Droge, B. and Polzehl, J. (1995). Model selection, transformations and variance estimation in nonlinear regression. Discussion Paper No. 95-52, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (Appeared in Statistics 33 (1999), 197-240.)
  • Bunke, O., Droge, B. and Polzehl, J. (1995). Splus tools for model selection in nonlinear regression. Discussion Paper No. 95-73, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (Appeared in Computational Statistics 13 (1998), 257-281.)
  • Bunke, O., Droge, B. and Polzehl, J. (1995). Selection of models and confidence intervals in heteroscedastic nonlinear regression. ASA 1995 Proceedings of the Biopharmaceutical Section, 233-238.
  • Droge, B. (1996). Orthogonal series regression estimation: projection vs. shrinkage. Discussion Paper No. 96-30, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (A revised version appeared under the title "Minimax regret analysis of orthogonal series regression estimation: Selection versus shrinkage" in Biometrika 85 (1998), 631-643.)
  • Droge, B. (1997). Asymptotic optimality of full cross-validation for selecting linear regression models. Discussion Paper No. 97-5, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (Appeared in Statistics & Probability Letters, 44 (1999), 351-357.)
  • Bunke, O., Droge, B. und Schwalbach, J. (2000). Die relative Bedeutung des Einflusses von Firmen- und Industriezweigeffekten auf den Unternehmenserfolg. Discussion Paper No. 101, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (A revised version appeared under the title "Die relative Bedeutung von Firmen- und Industriezweigeffekten für den Unternehmenserfolg" in Zeitschrift für Betriebswirtschaft 72 (2002), 275-294.)
  • Brenner, S., Bunke, O., Droge, B. and Schwalbach, J. (2001). The relative importance of group-level effects on the performance of German companies. Discussion Paper No. 95, Sonderforschungsbereich 373, Humboldt-Universität, Berlin.
  • Droge, B. (2002). Statistical methods of model selection and their application to economic problems. Proceedings of the Alumni Summer School "Applied Mathematics", Humboldt-Universität, Berlin, July 2002 (18 pp.).
  • Droge, B. (2002). On the minimax regret estimation of a restricted normal mean, and implications. Discussion Paper No. 81, Sonderforschungsbereich 373, Humboldt-Universität, Berlin.
  • Droge, B. (2003). Asymptotic properties of model selection procedures in linear regression. Discussion Paper No. 28, Sonderforschungsbereich 373, Humboldt-Universität, Berlin. (Appeared in Statistics 40 (2006), 1-38.).
  • Droge, B. (2005). Panel Data Analysis. Lecture notes, Wirtschaftswissenschaftliche Fakultät, Humboldt-Universität, Berlin.
  • Droge, B. and Uhl, B. (2006). Minimax regret estimation in the linear regression
    model under possible heteroskedasticity. Manuscript.
  • Karaman Örsal, D. and Droge, B. (2009). On the Existence of the Moments of
    the Asymptotic Trace Statistic. Discussion Paper 2009-012, Sonderforschungsbereich 649, Humboldt-Universität, Berlin.
  • Karaman Örsal, D. and Droge, B. (2009). Panel Cointegration Testing in the
    Presence of a Time Trend. Discussion Paper 2009-005, Sonderforschungsbereich
    649, Humboldt-Universität, Berlin.