Author: Tom Starke

AI, Quantitative Analysis and Data Science Solutions for Finance and Manufacturing.

Quant Basics 8: Bootstrapping and Response Surface

In the last section we investigated how the strategies we’ve selected from our train-test cluster were distributed in the parameter plot. We saw that they form a dense cluster in that plot which indicates that the PnL’s we see are not a result of overfitting since we would expect them to be more randomly distributed.…
Read more

September 20, 2017 0

Quantopian Advanced Workshop

Presenters: Dr Tom Starke & Michelle Lin The workshop taught how to use the pipeline API, which enables you to apply your trading ideas to massive universes of stocks and futures while reducing the performance penalty through lazy evaluation. Next we showed how factor models work. In the US alone there are currently more the…
Read more

September 18, 2017 1


Dr Tom Starke will give a presentation on the 29 Sept with the title: Demonstration of Machine-Learning Based Strategy Parameter Selection in Python Abstract System Parameter Permutation (SPP) has been a hot topic in quantitative trading in the past few years. However, for most people this is still quite an abstract concept that is challenging…
Read more

September 5, 2017 0


Dr Tom Starke and Michelle Lin will be running the Quantopian Advanced workshop on 16 Sept 2017. Here are some information on it. For more information, please follow the link. Cross-sectional factor modeling is widely accepted by academics and industry practitioners alike as a general and consistent way to model and understand equity markets. We…
Read more

September 3, 2017 0

Monte Carlo Options Pricing in Two Lines of Python

This is an old video that I produced sitting on my bed in the morning in order to learn how to make basic Youtube videos. Nevertheless, I became quite popular. I apologise for looking a bit rough. Hope you enjoy it regardless. Here is the code: If that seems a bit complex, check out the…
Read more

September 1, 2017 0

Quant Basics 7: Identify Best Parameters

In the previous post we used unsupervised machine learning to identify the best cluster of train-test results. The question is: is that cluster meaningful? If the parameters of the moving averages for the cluster are scattered evenly over the parameter space, this cluster may not be very meaningful. On the other hand, if the parameters…
Read more

August 31, 2017 0

Quant Basics 6: Start With Machine Learning

In the previous section we ran a parameter sweep over the train and the test set of our strategy and looked at the average PnL. In this post we will start with system parameter permutation (SPP) in order to improve the performance of our system without falling prey to data mining bias. Remember that we…
Read more

August 28, 2017 0

Deep Learning for Trading: LSTM Basics for Pairs Trading

We will explore Long Short-Term Memory Networks (LSTM networks) because this deep learning technique can be helpful in sequential data such as time series. As its name suggests, it can ‘remember’ previous observations, which wouldn’t be so necessary in non-sequential data, but especially helpful for time series data like the financial markets. This is the…
Read more

August 27, 2017 2

Quant Basics 5: Parameter Sweep

Introduction In the last section we ran a single backtest. However, for our strategy to work we should optimise our strategy parameters. If we blindly run through a large set of parameters and then pick the best one we are very likely to fall for an issue called Data Mining Bias. This means that if…
Read more

August 25, 2017 0

Quant Basics 4: Analysing A Single Backtest

In the previous posts we have downloaded market data, developed a vectorised backtest and calculated PnL, Sharpe ratio and drawdown. In this post we will set up, run and analyse a single backtest. This is the basis for running parameter sweeps and optimisations with hundreds or thousands of backtests. So, let’s set some important parameters…
Read more

August 24, 2017 5