Discussing the article: "Reimagining Classic Strategies (Part VI): Multiple Time-Frame Analysis"
Thank you for your step by step and detailed analysis. This is really helpfully as a template for those unfamiliar with data analysis to work through . the result is fascinating (and unexpected), Does this mean we should be testing other parameters as well e.g. volume, spread Previous day action or is that more overfitting. Thanks for the template I now have the tools to check it myself :)
linfo2 #:
Thank you for your step by step and detailed analysis. This is really helpfully as a template for those unfamiliar with data analysis to work through . the result is fascinating (and unexpected), Does this mean we should be testing other parameters as well e.g. volume, spread Previous day action or is that more overfitting. Thanks for the template I now have the tools to check it myself :)
Hey Neil, I'm glad you find this useful. Thank you for your step by step and detailed analysis. This is really helpfully as a template for those unfamiliar with data analysis to work through . the result is fascinating (and unexpected), Does this mean we should be testing other parameters as well e.g. volume, spread Previous day action or is that more overfitting. Thanks for the template I now have the tools to check it myself :)
I'd like to believe that you're going in the right direction, we should definitely test other parameters, it's worth a check.
One way to overcome overfitting is to use larger datasets. That way we will have large training and validation sets that give a faithful representation of the broad market behavior, however this also takes more time to compute and therefore I avoid practicing it in the articles, my current laptop doesn't have the right resources for such a task
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Check out the new article: Reimagining Classic Strategies (Part VI): Multiple Time-Frame Analysis.
In this series of articles, we revisit classic strategies to see if we can improve them using AI. In today's article, we will examine the popular strategy of multiple time-frame analysis to judge if the strategy would be enhanced with AI.
In this article, we revisit a well-known strategy of multiple time-frame analysis. A large group of successful traders around the world hold the belief that there is virtue found in analyzing more than one time-frame before making investment decisions. There are many different variants of the strategy. However, they all tend to hold the general belief that whichever trend is identified on a higher time-frame, will persist on all time-frames lower than it.
So for example, if we observe bullish price behavior on the daily chart, then we would reasonably expect to see bullish price patterns on the hourly chart. This strategy also extends the idea further, according to the strategy, we should add more weight to price fluctuations that are aligned to the trend observed on the higher time-frame.
In other words, returning to our simple example, if we observed an uptrend on the daily chart, then we would be biased more towards buying opportunities on the hourly chart, and we would reluctantly take positions opposing the trend observed on the daily chart.
Generally speaking, the strategy falls apart when the trend observed on the higher time-frame is reversed. This is usually because the reversal will start out only on a lower time-frame. Recall that when using this strategy, little weight is attributed to fluctuations observed on lower time-frames that are contrary to the higher time-frame. Therefore, traders following this strategy would typically wait for the reversal to be observable on the higher time-frame. Therefore, they may experience a lot of volatility in price whilst waiting for confirmation from the higher time-frame.
Author: Gamuchirai Zororo Ndawana