Quality Control Data

M U S E N W E N

*** *** ******* ****** ***. *** Pasadena, CA 91106 951-***-****

*****.***@*****.***

EDUCATION

**/**** ********** ** ********** Riverside, CA

-

**/**** **. * ******* Statistics (supported by Dean's Distinguished Fellowship)

GPA: 3.85 (range over 20 Statistics courses)

Research : Nonlinear non-Gaussian time series modeling and mixture models

09/2004 Imperial College of Science and Technology, University of London, U.K.

- London

10/2005 M.Sc. Mathematics & Finance

09/2000 Beijing Institute of Technology (BIT) Beijing, China

-

07/2004 B.Sc. Mathematics & Applied Mathematics

EXPERIENCE

09/2006 Statistical Consulting Center, UCR

-

06/2007 Statistical Consultant

Analyzed and implemented Cox's survival models and quality control methods and

financial time series models for visiting clients of the statistical consulting

center

Worked along side faculty and grad researchers by providing comprehensive

statistical support and analysis

Synthesized, cleaned and pre-processed experimental data in order to facilitate

post statistical analysis

Wrote R and SAS codes to implement models for various projects ; effectively

communicated and presented end results to non-Statistics professionals

06/2006 Department of Statistics, UCR

-

08/2010 Teaching Assistant/Lecturer

Worked as teaching assistant for various upper level undergraduate Statistics

courses, including Probability Models, Regression, Time Series and Forecasting,

Bayesian Statistics, etc.

Lecturer for Statistics for Science and Engineering (class enrollment > 80)

SELECTED PROJECTS

Two new classes of nonlinear non-Gaussian multivariate time

series models

Proposed two new classes of nonlinear time series models; solved a series of

statistical problems involving model specification, parameter estimation (via new

ECM algorithm), goodness-of-fit test and forecasting

Applied the model to general marked point processes data (earthquake data, e.g.)

and high frequency economics data as an important illustration of the new models

Machine learning: Recurrent Reinforcement Learning (RRL)

algorithm

Investigated and analyzed a class of machine learning models - recurrent

reinforcement learning algorithm for FX trading Studied problems involving neural

network algorithms, performance function optimization and real time predictions

Others: survival analysis, quality control models, linear time

series models

Studied and analyzed and implemented the Cox's proportional hazard models, quality

control models for real problems from visiting clients of the Statistical

Consulting Center; Studied and developed the innovation algorithms for most linear

time series models' estimation and forecasting problems (written in R)

QUANTITATIVE SKILLS

Graduate Coursework: Probability Theory, Mathematical Statistics, Multivariate

Statistical Analysis, Time Series Analysis, Statistical Data Mining, Bayesian

Statistics, Statistical Computing, Nonparametric Statistics, Discrete Data

Analysis, etc.

Computer Skills: Proficient in R, SAS; Fundamental programming in C++ and Python;

Working knowledge with UNIX

PUBLICATION

Wen, M. and Lii, K.S. Multivariate MTD time series framework for marked point

processes, Computational Statistics & Data Analysis (Submitted, 2009)

Wen, M. and Lii, K.S. Multi-logit mixture autoregressive time series model: a

model for high frequency data, Journal of Applied Statistics (Submitted, 2010)

ACTIVITIES

Conference presentation - Joint Statistical Meetings (JSM), Washington, D.C.

(08/2009)

Conference presentation - NSF Conference on modeling high frequency data in

finance, Hoboken, New Jersey (07/2009)

Invited to review papers for Journal of Applied Statistics (2008-now) and Handbook

of modeling high frequency data (2010)

Contact this candidate