quant financial engineer
LATEST PROJECTS
To see more or discuss possible work let's talk >>
#Machine learning
#Random fluctuation
#Hyper-parameter optimization
# Model Confidence Set approach
# Covid-19 prediction
Hyper-parameter optimization is a thoroughly investigated discipline in the field of statistics and machine learning. In this paper I shed light on a serious hidden issues why most such optimization techniques fail. I propose a Model Confidence Set extension of grid search to react to these issues.
I have developed an MN-SEIR model to forecast Covid-19 confirmed cases in New York which incorporates contact tracing and isolation.
It turns out that the propozed hyper-parameter opt. technique outperforms grid search.
The method is extremely powerful in choosing profitable trading strategies with respect to the Sharpe-ratio.
#RelativeStrengthInvesting
#Carry-trade
#Momentum trade
An empirical study about the profitability of the famous momentum and carry trade via relative strengh investing. In this paper, I study FX markets as a natural laboratory for momentum and interest rate returns. The data covers the sample period from January 2005 to January 2018. As is turns out, significant positve return can be accessed by applying the aforementioned models. Numerically speaking, the so-called double sorting method in the FX market yields to a 11.27% annual return at a 10% significance level.