Visit our COVID-19 site for latest information regarding how we can support you. For up to date information about the pandemic visit www.sacoronavirus.co.za.

bs-regular
bs-extra-light
bs-light
bs-light
bs-cond-light-webfont
bs-medium
bs-bold
bs-black

Community


Share knowledge. Ask questions. Find answers.

Online Share Trading

Engage and learn about markets and trading online

Trading models - John

Reply
MenuOption
Frequent Contributor
John, if I recall you had software trading models built a few years back. You still using it? I finally got mine going using a number of trading strategies. Some good results, but also some not so good.
0 Kudos
3 REPLIES 3
john_1
Super Contributor
yes Ihave 2 systems.one for the alsi one for share trading.. Share tdading is still been refined. but it works and it worked very well through the crash this time a year ago. so it will work in calmer times. what can I help you with.
0 Kudos
MenuOption
Frequent Contributor
Great. What Ive done is implement a number of the "standard" trading techniques - MACD, RSI, trailing stoploss, etc. and evaluated the performance of these systems against each other if optimized - per share. Then I developed a number of additional strategies that's not part of the normal litterature. Shares are evaluated against each other using a set of metrics that include trading history and fundamentals (PE, Div, etc) to determine the set of tradable shares.
Before the crash I did not want to use the system because of fears of how it will perform during such a corrections. But now it is ready for a test-run.
Overall it does very well - but for specific shares the system does not perform all that well - AGL for example - I can manage only 10% pa consistently.
So was wonderring if your system does better. Im looking at ways of improving the system for the shares that it does not trade well.
The verbose responce mostly to give you background. Thanks.
0 Kudos
john_1
Super Contributor
sounds like it does not like volitility..thats ok then just add in some kind of tolerence for that and see how you do....the more optimized the more likely it will fail if the parameters change..again thats fine as long as you know what might cause a failure...this data on AGL might turn out to be very key in this regard..
0 Kudos