Banking at the Crossroads Essays on Machine Learning / Applied Data Analytics, Asset Encumbrance / Bail-In, Sustainability and Resource Availability

Banking at the Crossroads Essays on Machine Learning / Applied Data Analytics, Asset Encumbrance / Bail-In, Sustainability and Resource Availability

Learning Algorithms, and current developments – Banking at the Crossroads Essays on Machine learning / Applied Data Analytics, Asset Encumbrance / Bail-in, Sustainability and Resource Availability Dissertation zur Erlangung des akademischen Grades Doctor rerum politicarum an der Fakultät Wirtschaftswissenschaften Technische Universität Dortmund Dipl. Wi-Ing. Joachim Erhardt London, December 2017 Acknowledgments This PhD thesis was supported by Landesbank Baden-Württemberg. I am grateful to Professor Dr. Peter N. Posch for his ongoing encouragement and valued advice. I would like to express thanks to my team, particularly to Dr. Rainer Dürr, Tarek Hard (PhD) and David West for their support and modelling work over the years. I am thankful for having had the opportunity to supervise the master theses of Mark Wolters, Johannes Luebbers, Bernd Schumacher and Patrick Kosa. Last but not least, I would like to thank my wife and especially our daughter for their mostly unwavering patience. ii Contents Contents ................................................................................................................................... iii 1. Introduction ........................................................................................................................ 1 2. Efficiency Impact through Data Analytics ......................................................................... 4 2.1. Introduction ................................................................................................................. 4 2.2. Machine Learning – Elements of Choice .................................................................... 5 2.3. Application – Market Trends .................................................................................... 27 2.4. Application – Institutional Customer Behaviour ...................................................... 62 2.5. Summary and Conclusion - the use of Learning Algorithms in Finance .................. 76 3. Bail-In and Asset Encumbrance: Implications for Banks’ Asset Liability Management . 81 3.1. Introduction ............................................................................................................... 81 3.2. Data Description ........................................................................................................ 83 3.3. Simulation Setup ....................................................................................................... 90 3.4. Effect of Asset Encumbrance on Costs ..................................................................... 94 3.5. Conclusion ............................................................................................................... 100 4. Sustainability, Green Bonds and Ethics ......................................................................... 101 4.1. Introduction ............................................................................................................. 101 4.2. Business Case for Investing under ESG – Economic Return.................................. 103 4.3. Conclusions on ESG - and Social Return................................................................ 107 4.4. Green Bond Solution ............................................................................................... 110 5. The Impact of Commodity Finance on Resource Availability ....................................... 113 5.1. Introduction ............................................................................................................. 113 5.2. Empirical Analysis .................................................................................................. 114 5.3. Conclusion ............................................................................................................... 117 6. Concluding Summary ..................................................................................................... 119 7. Literature ........................................................................................................................ 124 8. Appendix ........................................................................................................................ 8-1 8.1. Appendix on Elliott Wave Theory .......................................................................... 8-1 8.2. Appendix on Trend Prediction ................................................................................ 8-7 8.3. Appendix on Institutional Customer Behaviour.................................................... 8-40 iii List of Tables Table 1: Classic Feature set ..................................................................................................... 32 Table 2: The Landmark Feature set ......................................................................................... 34 Table 3: List of Residuals and ‘Landmark light’ Features ...................................................... 44 Table 4: Confusion Matrix – NKY, Multi-asset with Granger Causality and Classic Features. .................................................................................................................................................. 45 Table 5: Prediction statistics – Overview across models for NKY, DAX, and SPX Index. .... 47 Table 6: Confusion Matrix – UKX, Multi-asset with Granger Causality and ‘Residuals & Landmark light’ Features. ........................................................................................................ 50 Table 7: Confusion Matrix – DAX, Multi-asset with Granger Causality and ‘Residuals & Landmark light’ Features. ........................................................................................................ 50 Table 8: Confusion Matrix – SPX, Single-asset with Landmark Features. ............................. 52 Table 9: Confusion Matrix – DAX with XGBoost, Multi-asset with Granger Causality and Residuals. ................................................................................................................................. 54 Table 10: Confusion Matrix – DAX with Random Forest, Multi-asset with Granger Causality and Residuals. .......................................................................................................................... 54 Table 11: Comparison of Prediction Statistics according to Customer Activity Ranking. ..... 72 Table 12: Funding composition of the bank sample. ............................................................... 85 Table 13: Simple average over-collateralisation levels for covered bonds. ............................ 86 Table 14: The level of asset encumbrance (AE) and deposits. ................................................ 87 Table 15: Approximated funding curves based on simple Z spreads. ..................................... 89 Table 16: Shown are the percentage share of covered bonds (Covered) based on a 2% asset loos scenario............................................................................................................................. 95 Table 17: Shown are the percentage share of covered bonds (Covered) based on a 10% asset loss scenario. ............................................................................................................................ 97 Table 18: Shown are the percentage share of covered bonds (Covered) based on a 20% asset loss scenario. ............................................................................................................................ 98 Table 19: Fixed Effects Regression of Import Changes, worldwide ..................................... 116 Table 20: Fixed Effects Regression of Import Changes, European Union ............................ 117 Table 21: Calculation of reversal prices based on Fibonacci multiples................................. 8-1 Table 22: Assigned probabilities for alternative multiples, ................................................... 8-1 Table 23: Order of preference for wave patterns in ‘up- and down-trends’. ......................... 8-3 Table 24: EW-Features used as modelling set ....................................................................... 8-5 iv Table 25 (cont.): EW-Features used as modelling set ........................................................... 8-6 Table 26: List of Leading Markets – out of 300 assets - based on Granger Causality test: .. 8-7 Table 27: Confusion Matrix NKY - Single time series with ‘Classic Features’. ................ 8-14 Table 28: Confusion Matrix NKY – Single time series with ‘Landmarks’. ........................ 8-15 Table 29: Confusion Matrix NKY – Single time series with ‘EW Features’. ..................... 8-16 Table 30a, 30b: Confusion Matrix & Performance Statistics NKY – Multi-asset with Granger and ‘Landmarks’. ................................................................................................................. 8-16 Table 31a, 31b: Confusion Matrix & Performance Statistics NKY – Multi-asset with Granger and ‘EW-Features’. .............................................................................................................. 8-17 Table 32a, 32b: Confusion Matrix & Performance Statistics NKY – Multi-asset with Granger and ‘Residuals’..................................................................................................................... 8-18 Table 33a, 33b: Confusion Matrix & Performance Statistics NKY – Multi-asset without Granger and ‘Residuals’. ..................................................................................................... 8-19 Table 34a, 34b: Confusion Matrix & Performance

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    188 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us