The Walk-Forward Loop performs the back-testing, but how do we go about interpreting its results? The very nature of the walk-forward loop makes the test data set approach (test the performance on a left-out portion of the data) unattractive.
About a couple of years ago, I rolled out the flock package to help me synchronize R processes. I have used it ever since, but it wasn’t until recently that I found the time to move the source code to GitHub, and to add the package to CRAN.
The previous post described the high level architecture of a walk-forward forecasting for time series data. As a hands-on implementation – let’s apply a simple QDA classifier on the series discussed previously.
For quite some time now I have been using R’s caret package to choose the model for forecasting time series data. The approach is satisfactory as long as the model is not an evolving model (i.e. is not re-trained), or if it evolves rarely. If the model is re-trained often – the approach has significant computational overhead. Interestingly enough, an alternative, more efficient approach allows also for more flexibility in the area of model selection.
The previous post in this series, showed a way to identify trading opportunities. The approach I implemented used time series daily data to identify good entry points in terms of risk-reward. The natural next step is to try to make use of these opportunities using machine learning.
Is trading worth doing for living? Once in a while I am asked this question and, to me, going through this mental exercise makes sense. A lot of sense. So let me share my thoughts.
One approach to trading which has been puzzling me lately, is to sit and wait for opportunities. 🙂 Sounds simplistic, but it is indeed different than, for instance, the asset allocation strategies. In order to be able to even attempt taking advantage of these opportunities, however, we must be able to identify them. Once the opportunities are identified – we can try to explain (forecast) them using historical data.
On at least a couple of occasions lately, I realized that I may need Python in the near future. While I have amassed some limited experience with the language over the years, I never spent the time to understand Pandas, its de-facto standard data-frame library.