Last week we crossed the one year mark after president’s election. Both in calendar days, November the 8th, and in terms of trading days – 252 (the average number of trading days per calendar year). I wanted to do a brief recap before leaving it to rest.
When I wrote the original post, I wasn’t planning on writing a follow-up. Certainly not the week after. But what a difference a week can make in a dynamic system like the US stock market.
Coming back to markets and trading (after a while), the feeling has been that the markets, and the economy as a whole, are doing good. How good? Since I haven’t been following things closely, I had to do some forensics.
At least in my town. Sometime back I kind of decided to use Quantopian as the backtesting platform of my choice. QuantConnect was a close second best. Now, a few months later, Quantopian has decided to end the live trading. No choice, but to go back to the second choice.
Read More
Neural networks have been around for a while, but it’s fair to say that many successful practical applications use at least one convolutional layer. Naturally, convolutions make sense for time series, so I went and added a few to the Walk-Forward Analysis.
Read More
A previous post in this series, implemented the Walk Forward Loop on top of Microsoft’s CNTK. There was interest in a Google’s Tensorflow implementation, which seems to be the more popular framework in this domain, I decided to put what have already done with Tensorflow.
Read More
Backtesting – the pillar of trading and investing. I know what all the naysayers say, but with all due respect, they got this one wrong. 🙂 In finance, and in time series in general, history repeats itself. Natural phenomena follow patterns, so does crowd behavior. Earth temperatures are not going to rise forever, so wouldn’t the current bull market. Enough rambling though.
My readers know that my main tool for backtesting has been Tradelib, a Java library which I have developed and used over the past couple of years. Working on some new (or rather old) project, I decided to take another look what’s around, and the results surprised me.
Read More
In the previous posts in these series (here, here and here) I used conventional machine learning to forecast the trading opportunities. Lately however I have been trying to move more and more towards deep learning. My first attempt was to extend the walk-forward loop to support neural networks, the building blocks of deep learning.
The previous post introduced forecasting using multiple series and also suggested another form of improvement – namely filtering out low probability forecasts. Can we improve the forecasts by any of these approaches, and if yes, by how much?
Read More
