Research Assistant in Statistical Data Analysis at ZHAW
|Career Status:||Actively looking|
|Willing to relocate:|
|Willingness to travel:||Fairly willing to travel|
Currently I am looking for a permanent position of a (junior) data scientist / statistician in Switzerland.
I would be delighted to hear from you if you have suchlike vacancies.
• Background in Machine Leaning, Time Series Analysis and Mathematical Finance.
• Experience in many applied projects in statistical consulting with advanced use of R, Python and Matlab.
• Applications in production engineering, geostatistics, finance, business decision making and web application development (R Shiny)
2011 – 2014: ETH Zürich, MSc Statistics
Emphasis on advanced probability, time series analysis, multivariate statistics and machine learning.
MSc thesis “Parameter estimation for diffusion processe” (code in Matlab).
Supervisor: Prof. Dr. Hans-Rudolf Künsch
Used estimating functions approach in discretely sampled diffusions. Built a Monte-Carlo simulation, identified advantages and deficiencies of root-search algorithms and causes for asymptotic bias and inflated variance in the estimators.
2008 – 2011: University of London, BSc Mathematical Economics and Finance
Lead college: London School of Economics (LSE)
Emphasis on quantitative finance, mathematical economics, econometrics, and optimization.
2007 – 2011: Higher School of Economics (Moscow), BSc Mathematical Economics
(Double-degree programme with University of London)
BSc thesis “Methods of pricing of European type options with heavy-tailed distribution of the underlying”. Defended with excellence.
Found strong sensitivity of options with log-Levy alpha-stable distributed underlying (Hurst-Platen-Rachev model) to stability index & volatility, which may render it worse than Black-Scholes. Solved numerical integration instabilities.
90% – Applied Research Projects with Industry Clients.
10% – Teaching courses in Machine Learning, Data Analysis and R programming.
R packages used: caret, dplyr, reshape, data.table, lattice, ggplot2, ggvis, shiny, manipulate, sp, igraph, knitr, mice, mvoutlier, randomForest, nnet, cluster, MASS, car, sandwich, roxygen2.
Python libraries used: numpy, scipy, pandas, lasagne, re, matplotlib.
Matlab toolboxes used: System Identification, Econometrics, Statistics and Machine Learning.
Data preprocessing, visualization and model calibration in the following projects:
• Pattern identification in physical/chemical data to find the optimal concrete mixture
(Generalized Linear Models, R)
• Designing a business recommender system to predict competitors’ strategic actions
(Multidimensional scaling and graph clustering, R)
• Fuel consumption optimization for marine vessels
(Generalized Additive Models, R)
• Predicting gas turbines failure from sensor data
(Multivariate time series with exogenous variables, Matlab)
• Pattern identification in passenger commuting with Web apps
(Python, pandas, numpy, regular expressions)
• (current) Detecting defects in loudspeakers production
(Convolutional Neural Networks, Python, lasagne)
• Received excellent faculty references upon completion of probation period
Teaching tutorials of graduate course on Stochastic Systems.
• Stochastic Differential Equations (univariate and multivariate)
• Time Series Analysis (SARIMA-GARCH)
• Kalman filtering
• Stochastic Optimal Control, Hamilton-Jacobi-Bellman equation and application to portfolio optimization
• Calibrating Hawkes self-excited point process model on high-frequency financial data
• Investigating behavior of non-parametric estimators of the branching ratio of the Hawkes process
• Monte-Carlo experiments in Matlab
- Time Series Analysis
- Machine Learning
- Mathematical Statistics
- Stochastic Processes
- Numerical Analysis
- Data Visualization
- Explorative Data Analysis
- Advanced Regression
- Generalized Linear Models (GLM)
- Generalized Additive Models (GAM)
- English, German, Russian