- This topic is empty.
-
Cheers Scott.
I’ve been working my way through the trade list to confirm all is clean….. so far so good, so the figures look legit.
This system is essentially the same system as the rotational system I’ve been building for the NDX100. The main difference is the index filter…….
This index filter is based on Mike’s idea of the dual MA crossover on the index. I’m calling it the “Rodwell Regime Filter”
I had explored/tested the dual MA on the index in the NDX100 build, but didn’t see a benefit with it for my system (over the standard 200MA index filter).
When I cut the system over to the ASX100 to test robustness, initially the equity curve & DD was a bit ‘choppy’….. Nick suggested Mike’s approach, which I had already tried on the NDX100 & kind of disregarded it for the ASX100 build (I have no idea what Mike is using for his lookback periods etc), but I used his principal for the dual MA crossover and ‘searched’ for parameters that worked for my system…. and what you see in the reports is what come out
There were actually a few parameters, i.e. lookbacks (in the same ballpark) which appeared to work quite nicely (which is what you want from a robustness perspective).
So I guess the learning from my perspective is, just because I ‘tweak’ didn’t work on a system in one market, doesn’t mean the same tweak won’t work on the same system in another market.
Interestingly, using the “Rodwell Regime Filter” on the ASX100, if I were trading the system live ‘today’…. the index filter would still be UP/Green status, which means I’d still be invested in the market during this month of MARCH. On the flip side, I would’ve been out of the market at the end of September 2018 and would’ve been in ‘cash’ at the start of OCT during the OCT 2018 downturn. Probably a bit ‘recency bias’, looking back at these time periods, but it’s a process of getting an understanding/feel of how this index filter operates etc.
Cheers
LOL… might be a bit of a stretch taking naming rights for a MA cross over (but I’ll have it!)
A little bit of a white knuckle ride at the moment while a lot of momentum systems are switched off… may be this bounce will continue.
Re: your recency bias, yeah, my system skipped a lot of the carnage other experience back in Sept/ Oct but it relies on a bounce happening so it could get caught out. If the market chops around here for a while my MA’s will tighten right up and be sensitive to a downward movement (ie, switch the system off sooner rather than later).
In this image:
1. V shape sell off is no dramas
Then the market chops around and my MA’s tighten up
2. a few down days and its out…
Thought I’d list some of the resources I’ve used for my Rotational build(s), so that I and/or others can refer to it in the future….
I’ve also made a couple of ‘tweaks’ to my current ASX100 Rotation Build…..
* I’ve removed the indicator check on the stock for the ASX100 build. This indicator check worked/works very nicely on the NDX100 Rotation build…. but it was making very little difference on the ASX100 build (still active on the NDX100 build).
* In light of recent weeks….. I’ve added an Index ‘Kill Switch’ Filter… e.g. if the Index (XAO.au) drops X% over the past Y days, then the Index filter turns from GREEN to RED and we exit all positions @ the end of the month. Testing so far shows it doesn’t activate all that often (outside of the current Index Filter) but in times like we are seeing at the moment…. any rapid drops in the INDEX will see us exit if the Normal Index Filter lags at the end of the month. So in effect, I have the Dual MA Index Filter and the “Kill Switch” Index Filter working along side each other…. The “Kill Switch” Index Filter, provides a little more in the “comfort factor” stakes…
– Said’s Ranking with ROC/ATR combined.
https://edu.thechartist.com.au/Forum/trading-system-idea-s/262-said-s-ranking-method.html#4124– Multiple Momo ROC Combinations
– Low Vola MOMO
https://alvarezquanttrading.com/blog/low-volatility-stocks-20-cagr-portfolio/– Rotation Readers Ideas
https://alvarezquanttrading.com/blog/sp500-monthly-rotation-readers-ideas/– GTAA
https://forum.amibroker.com/t/gtaa-timing-model-rotational-strategy/4090– Technical Ranks
https://alvarezquanttrading.com/blog/stockcharts-technical-rank-sctr-rotation-strategy/– Nick’s MOMO Video….. It is a must view:
https://www.tradelongterm.com/relative-momentum-video/– Momo Weighting.
– 200MA Distance Code
Code://https://forum.amibroker.com/t/indicator-distance-between-ma-200-and-price/17608/2
//
// Distance from MA 200 Day
ma200 = MA(Close, 200);distancia = (Close-ma200)/Close * 100;
cero = IIf(Close >= ma200, colorGreen, colorRed);
Plot( distancia, “Distance From 200DAY MA”, cero, styleHistogram | styleThick, Null, Null, 0, 1, 5 );
Example of the 200MA Distance
Scott McNab wrote:Glen Peake wrote:* I’ve removed the indicator check on the stock for the ASX100 build. This indicator check worked/works very nicely on the NDX100 Rotation build…. but it was making very little difference on the ASX100 build (still active on the NDX100 build).2c worth
I have tended to leave these in …. my rotational systems, even when tested over longer time frames, do not have a huge number of trades…and if it has proven beneficial for one data set while not harming the other then I find it hard to conclude that it may not become useful in future years…hhhhhmmm you might have a point there Scott… your 2c worth, warrants further review… as it does seem to weed out some of the stocks losing their MOMO… it may not make a huge difference in terms of ‘large’ chunks as on the NDX100….. but it’s the ‘one percenter’s (1%er’s) that can make a slight difference.
Thanks for the nudge….
With Stock Filter:
Without Stock Filter:
Let’s face it…. I’ll probably have at least a few months more of testing, before the INDEX Filter goes green again anyway
- You must be logged in to reply to this topic.