Sat 22 September 2018
Predicting future stock price movement is known difficult due to low signal-to-noise ratio and non stationary price distribution. Off-the-shelf successful ML algorithms often end up giving you disappointed results. Indeed, a lot of ML quant hedge funds show up and disappear every year.
To deal with this difficult problems, we ...
Sun 26 August 2018
PyTorch Implementation for Integrated Gradient
Fri 22 June 2018
Blockchain is one of the hottest technologies as well as AI/Machine Learning. Indeed, a lot of startup are working in this fields. Besides that compared to another hot topic, Deep Learning, Blockchain is still in early stage and there is space you can get in [1]. I know that ...
Sun 11 March 2018
Choosing a good set of hyperparameters is one of most important steps, but it is annoying and time consuming. The small number of hyperparameters may allow you to find an optimal set of hyperparameters after a few trials. This is, however, not the case for complex models like neural network ...
Thu 01 March 2018
Blogging is one of the fantastic ways to demonstrate your projects and help you understand stuff in more depth. Especially, I believe that blogging helps you land a job more efficiently. Even if you are not looking for a new position, writing articles you are working on would be the ...
Thu 26 October 2017
Applying hypothesis test for financial time series data
Sun 26 March 2017
Data exploration is important. Before jumping into analysis, you will get some intuition as to how to use data and establish more sophisticated modeling. In some cases, how to use data improve even models robust to high dimensional input like Deep Learning. In this article, we work on data exploration and visualization for the house price data from a kaggle competition
Sat 28 January 2017
Random forests and its application in feature selection.
Fri 13 January 2017
Review for three popular vector representations and work on predicting movie evaluation throuth review text.
Sat 07 January 2017
Reinforcement learning has recently been succeeded to go over the human's ability in video games and Go. This implies possiblities to beat human's performance in other fields where human is doing well. Stock trading can be one of such fields. Some professional In this article, we consider application of reinforcement learning to stock trading. Especially, we work on constructing a portoflio to make profit. Since portfolio can take inifinite number, we tackle this task based on Deep Deterministic Policy Gradient (DDPG). The behavior of stock prices is konwn to depend on history and several time scales, which leads us to use multiscale CNN for Q and actor network. We show that our algorithm has better performance than monkey trading(taking actions at random).