- Joined
- 28 December 2013
- Posts
- 6,392
- Reactions
- 24,319
I think avoiding answering 3, 4, and 5 says a lot.
Sarcasm. Sledging. It's all the same Phil. You're as guilty as me. Goodbye.
" I only buy when the "Percentage of advancing positions in the watchlist exceeds 50%" & sell when the percentage of advancing positions in the watchlist is under 25%"
I'm a little dumb. Can you elaborate on "advancing positions"? Is that like signals, aka positions, generated by the system, or do you mean "advancing" symbols within the watchlist. So in the XAO, if 50% of symbols are rising, then we're good to buy?
Are you trading just the watchlist, or is the watchlist selected to be representative of the overall market (ie your watchlist is what you believe is the market leaders so you expect them to rise before the market rises)?
I guess what I'm asking is, is the watchlist just whatever you are interested in trading
Thank you for the explanation of the PercentageUp filter.
Breadth filters have merit in some forms, but are not necessarily robust across a variety of strategies. Indeed, when I test breadth filters, I see instances where they can work well, yet on different strategies, they fall over. Not a good sign of robustness.
Unfortunately with just 1-year of backtested data, no stress testing data and no historical constituent data, it's difficult to assign any credibility to what you're showing. My only guess, based on my limited experience, is that there is data mining at work.
Until further evidence is provided I have nothing more to add to this discussion.
Good evening monkey_zu:So, I can make a list, any constituents (of XAO), any size of list, and if half of them are rising it will correlate to a rise in the overall market? Seems something is missing here
So, I can make a list, any constituents (of XAO), any size of list, and if half of them are rising it will correlate to a rise in the overall market? Seems something is missing here
Skate is saying that his percentageup filter is based on checking if more than 50% of his watchlist are trending higher, and that this same filter is very accurate in signalling the start of an upswing of the XAO. But the watchlist is not selected as the market leaders, and are just what he is interested in trading. Just seems to me the size and makeup of the list must have some impact on how well it correlates to the start of a market swing. Or I'm missing something, which is very possible.Good evening monkey_zu:
Welcome to Aussie Stock Forums. Not entirely sure what you are trying to say.
Have read of this and see how you go.
Watchlist: Definition, Purpose, and How To Create One
A watchlist is a list of securities being monitored for potential trading or investing opportunities.www.investopedia.com
Hope this helps.
Have a very nice evening.
Kind regards
rcw1
Thanks I'll have a look@monkey_zu the "PercentageUp Filter" that I've been discussing is essentially a market timing filter that helps me time my entry into the market. To gain a better understanding of how this filter works, I suggest researching the filter using the [Search] function using the keyword "PercentageUp" as posted by Skate.
There, you'll find a wealth of information, including a detailed discussion of the filter & my explanations of its mechanics. While there may be some repetition of information, it will provide a more comprehensive understanding of the filter's benefits & limitations. I appreciate your interest & willingness to learn more about my approach.
View attachment 155025
Skate.
I did a search. Not really any more info under 'percentageup'. So the watchlist you are using in your filter is the same as the watchlist you are trading? Does making a watchlist from a random selection from the XAO universe change the results?Thanks I'll have a look
I did a search. Not really any more info under 'percentageup'. So the watchlist you are using in your filter is the same as the watchlist you are trading? Does making a watchlist from a random selection from the XAO universe change the results?
"I understand your scepticism about the validity of my "PercentageUp Filter" due to the limited backtested data I've provided"
So post a 20-yr backtest. And do so with the correct data.
Can't see why it's so difficult.
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?