Teaching

University of Illinois at Urbana-Champaign:

• STAT 432: Basics of Statistical Learning

Supervised Learning: Linear regression & model selection, Ridge regression & Lasso, Classification and logistic regression, K-nearest neighbors & curse of dimensionality, Decision trees & random forests, Support vector machines & kernel trick. Unsupervised Learning: Principal component analysis, K-means & hierarchical clustering. Statistical concepts such as Bias-variance trade-off and resampling methods.

Columbia University:

• STAT GR5265/GU4265: Stochastic Methods in Finance

Discrete time models: Asset pricing with binomial model, American option pricing and optimal stopping problems, Utility maximization, Trinomial models and incomplete markets. Continuous time models: Review of stochastic calculus tools (Ito integral, Ito's formula, Girsanov theorem), Black Scholes model, formula and PDE, Pricing of European derivatives and exotic options, Stochastic volatility models, Interest rate models, Brief introduction to stochastic optimal control and Mean field games. 

• STAT GR5263/GU4263: Statistical Inference & Time Series Modeling 

Stationary time series, detrending and deaseasonalizing time series, ARMA processes: estimation and prediction, ARIMA and SARIMA models for nonstationary time series, Conditional heteroscedastic models (ARCH-GARCH) for volatility modeling. 

Princeton University:

• ORF 505: Statistical Analysis of Financial Data

Heavy tailed distributions and copulas. Simple and multiple linear regressions. Nonlinear regression. Non-parametric regression and classification. Neural networks: Multi Layer Perceptron, Convolutional NN, Recurrent NN (LSTM & GRU), Generative Adversarial NN.

• ORF 515: Asset Pricing II: Stochastic Calculus and Derivatives

Pricing and hedging of advanced derivatives, including topics such as exotic options, greeks, interest rate and credit derivatives, as well as risk management. Basics of stochastic calculus necessary for finance.

• ORF 405: Regression and Applied Time Series 

Linear, nonlinear, and nonparametric (kernel and projection pursuit) regression. Neural networks, convolution networks, deep learning with Tensor Flow and Keras. Classical linear time series models (AR, MA, ARMA) in univariate and multivariate settings.

• ORF 335: Introduction to Financial Mathematics 

Notions of arbitrage and risk-neutral pricing in discrete time, specific models such as Black-Scholes and Heston in continuous time, and calibration to market data. Credit derivatives, the term structure of interest rates, and robust techniques in the context of volatility options.

Boğaziçi University:

• IE306: Systems Simulation 

Basic concepts of discrete-event simulation modeling/analysis. Event-scheduling versus Process-interaction approach. Random number and random variate generation; inverse transformation and other selected techniques. Input data analysis and goodness of fit tests. Specific computer simulation languages. Analysis of simulation output and model validation.