Thanks JakoI will keep those suggestions in mind and test & use...

First 2612 Last

Share
Profile

Follow
sharks37

8,720 Posts.

81

08/06/20

16:57

Post #: 45095910
Share

Originally posted by Jako8557: ↑

Cool, that is a nice improvement in the results sharks,

I find the "close back above the high of a potential breakdown bar" a really good measurement of underlying positive sentiment (or a change back to a positive sentiment) in a financial instrument (almost any instrument), and I use it more and more often.
For me, judging the underlying sentiment in the market is really important, so I am constantly attempting to measure and confirm the underlying sentiment, especially when a change in sentiment maybe about to occur.

And just thinking about your model some more, perhaps keep in mind that 'stocks' in particular (especially larger cap stocks), will generally trade in harmony with their sector or parent index, so both MQG and CSL will also move in-line with the parent index (XJO) to a fair degree, no matter what the stocks individual sentiment is (generally this means that if the parent is weak on a particular day or week, a stock with a strong underlying sentiment will trade less weak than the parent - and if the parent is strong, the strong stock will be as strong or stronger than the parent on a particular day or week).

And including the sector index as well, may bring another level of complexity, but can be important when the sector trading against the parent to some degree, and the sector is positive and the parent is neutral or negative (or vice versa). The Gold sector comes to mind here as a classic case, but even more recently the financials have been a constant drag on the market, and the healthcare sector is currently acting weaker than the parent.

Anyway what I am getting at is - if you could limit the stock entries to when the stock and the parent Index are both trading in some harmony (or at least have the index in a favourable position when an entry is triggered in the stock), I expect the results may improve further.

And just wondering, are you able to include expected support and resistance zones in the model ??

cheers

.

Expand

Thanks Jako

I will keep those suggestions in mind and test & use if I can My key limitation in adding complexity to a model to improve its performance is the limit to which I can code that complexity (my intent here is to code my model and trade it in 100% auto mode, which is how I operate my FX model portfolio). Because of that limitation I have found the approach that works the best for me is to build a bunch of simple models which when combined in a portfolio provide the diversification needed to meet my target portfolio performance objectives (return, drawdown, etc).

Cheers, Sharks
3 Upvote

Reply