Sug­ges­ted top­ics for Mas­ter-Thes­is

Master-Theses in Financial Econometrics and Financial and Economic Data Science, SS 2023

 

Subareas and details

1. subarea I: Basic time series econometrics

I.a) Parametric, semi-parametric models for economic, financial or environmental time and their applications. Forecasting economic, financial or environmental time using semiparametric ARMA models possibly with seasonal component (topics based on M.184.4451 and Chapter 2 of W5453, the main package for this subarea is "smoots", x12, DeSeaTS).

I.b) Forecasting economic, financial or environmental time using semiparametric long-memory time series models (see M.184.5453, Ch3, coming, R package is "esemifar")

 

2. subarea II: Advanced Financial Econometrics

II.a) Modeling of high-frequency financial data (see Chapter 5 of M.184.5453)

II.b) Parametric and semiparametric long memory GARCH models and their application in quantitative risk management (see Chapter 6 of M.184.5453)

II.c) Parametric and semiparametric multivariate GARCH models and their application in quantitative risk management (see Chapter 7 of M.184.5453)

 

3. subarea III: Deep Learning with application to Financial and Economic Data:

a. Combination of deep learning with shortmemory (SM) GARCH models

b. Combination of deep learning with semiparametric GARCH models

c. Combination of deep learning with longmemory (LM) GARCH models

d. Combination of machine learning with semiparametric (LM) GARCH models

e. Deep learning with application to economic, business or environmental data.

 

Remark 1. The prerequisite for writing a Master Thesis at our Chair is that you have visited M.184.5451. If you still not yet have visited it but are going to visit this module in SS 2023, please indicate that you have visited W5333. This will be considered as an alternative to W5451.

Remark3. Topics Subarea II are suitable for those who are visiting the advanced time series module M.184.5453. If you are going to choose topics from Subarea III, the prerequisite is M.184.5452. You can also learn those topics yourselves.

Remark 3. Topics with BWL-focuses, e.g. studies on the behaviors of important German firms on the financial market, can be provided,

 

Master-Theses in Financial Econometrics and Financial and Economic Data Science, WS 2023/24

 

Subareas and details

1. subarea I: Basic time series econometrics

I.a) Parametric, semi-parametric models for economic, financial or environmental time and their applications. Forecasting economic, financial or environmental time using semiparametric ARMA models possibly with seasonal component (topics based on M.184.4451 and Chapter 2 of W5453, the main package for this subarea is "smoots", x12, DeSeaTS).

I.b) Forecasting economic, financial or environmental time using semiparametric long-memory time series models (see M.184.5453, Ch3, coming, R package is "esemifar")

2. subarea II: Advanced Financial Econometrics

II.a) Modeling of high-frequency financial data (see Chapter 5 of M.184.5453)

II.b) Parametric and semiparametric long memory GARCH models and their application in quantitative risk management (see Chapter 6 of M.184.5453)

II.c) Parametric and semiparametric multivariate GARCH models and their application in quantitative risk management (see Chapter 7 of M.184.5453)

II.d) Application of some most recently proposed long memory volatility models

II.e) Stationery and trend-stationary dual long memory processes in finance and economics

Subarea III: Deep Learning with application to Financial and Economic Data:

a. Combination of deep learning with short-memory (SM) GARCH models

b. Combination of deep learning with semiparametric (SM) GARCH models

c. Combination of deep learning with long-memory (LM) GARCH models

d. Combination of machine learning with semiparametric (LM) GARCH models

e. Deep learning with application to economic, business or environmental data.

 

Remark1. The prerequisite for writing a Master Thesis at our Chair is that you have at least visited one of M.184.4451, M.184.5451 (or equivalently M.184.5333 by Prof. Dr. Oliver Müller), M.184.5452 or M.184.5453. If you still not yet have visited one of those modules, but are visiting M.184.4451 or M.184.5452 just now, please indicate this in your application for the allocation of your Master thesis on the central allocation system.

Remark3. Topics Subarea II are suitable for those who are visiting the advanced time series module M.184.5453. If you are going to choose topics from Subarea III, the prerequisite is M.184.5452. You can also learn those topics yourselves.

Remark3. Topics with BWL-focuses, e.g. studies on the behaviors of important German firms on the financial market, can be provided,

Remark4. The choice of further advanced Econometric/Data Science topics, e.g. those in reinforcement learning or modeling of spatial data, is encouraged, provided that you have the previous knowledge on those topics. If this is your case, please discuss it with me in advance.

Chinese Page

对外办公时间 (2014/2015 冬季学期)

星期一 11:15 - 12:15 (或预约)

(假期中均需预约)

 

 

秘书

Felicitas Wax
Room: Q4.107
Tel: +49.5251.60-5003
Email: felicitas.wax@uni-paderborn.de

 

简历

教育背景: 北京师范大学数学系 (本科)

北京农业大学 (现中国农业大学) 经济管理学院 (硕士)

康斯坦茨大学 (德国) 统计学博士及德国大学教授资格

工作经历: 北京农业大学 (现中国农业大学) 经济管理学院

康斯坦茨大学数学与统计系及金融与计量经济研究中心

亨瑞瓦特大学 (英国) 精算与统计系

帕德博恩大学 (德国) 工商管理与经济学院 (2008-- )

 

 

主要课程(详见教学网页)

统计学,计量经济,时间序列分析,金融计量经济,应用计量经济,实证经济研究中的高级定量方法,

计量经济及统计的最新进展

 

 

主要研究方向

金融计量经济,时间序列分析, 非参数及半参数回归, 半参数金融模型,实证经济研究,实证金融研究,

计算机统计学, 金融大数据分析

 

主要科研成果(详见文章及论著)

非参数回归的带宽选择 (双平滑,1998,link; 简化双平滑,2009, link; 双平滑循环代入,2009,link)

数据驱动的时间序列分解(Berlin Method, 2000, 及其 循环代入算法,2013)

长记忆下的局部多项式(理论,算法,2002,异方差,2001,最忧收敛,2013)

半参数分数自回归模型 (SEMIFAR, 2002, link及其循环代入算法,2002)

半参数广义自回归条件异方差模型 (半参数 GARCH 模型=Semi-GARCH, 2004,link,及其高频扩展,2008)

半参数分数自回归-广义自回归条件异方差模型=SEMIFAR-GARCH (2007, link)

指数型半参数分数自回归模型(ESEMIFARSemi-FI-Log-ACD,2014)

半参数自回归条件持续期模型 (半参数ACD模型=Semi-ACD, 2014,link,及其详细算法,2013,pdf)

半参数不对称幂级数GARCH模型 (半参数APARCH模型=Semi-APARCH, 2013, pdf)

半参数GARCH模型族 (半参数EGARCH,半参数CGARCH等=Semi-EGARCH, Semi-CGARCH etc., 待发表).

高频金融数据空间模型(spatial MCGARCH,2013双重条件平滑,2015)

已实现核的数据驱动估计(data-driven estimation of realised kernels, 2015).

 

与国内大学联合培养的博士

郭志超 (中国农业大学, 2012), 现任教于北京工商大学经济学院

王江涛 (中南财经政法大学, 2014),现任教于华中师范大学经济与工商管理学院

 

 

部分学术活动

2013:组织并主持弟19届欧洲统计学会的长记忆时间序列分析的分组会

2013:在山东大学举办半参数金融时间序列模型专题讲座

2014:在本校成功举办 "非参数与半参数异方差及相关性模型" 国际会议

 

教研组成员

见我们的德文网页

Prof. Dr Yuan­hua Feng

business-card image

Prof. Dr. Yuanhua Feng

Ökonometrie & Quantitative Methoden

Econometrics, Financial Econometrics, Time Series Analysis, nonparametric regression

Write email +49 5251 60-3379

Sec­ret­ari­at

Sec­ret­ari­at

Re­search as­sist­ants

Shujie Li

Domin­ik Schulz

Oliv­er Kojo Ay­ensu

Aliyu Abubakar Musa

WHB

Thi Thu Huong Do