Diversified Trend Trading Approach

Looks like trade manager is working great. It bought correct amount of lots for 1% risk and it moves the SL:

[quote=Ramadas;10543558]Looks like trade manager is working great. It bought and it moves the Sl

I'll have another look at the Ramadas version today and make sure it is working this time

I got around to check correlation by price curve comparison and here it is.... for 30m.

EUR CAD = EUR AUD = EUR USD = EUR JPY



GBP CAD = GBP AUD = GBP USD = GBP JPY

EUR CAD = 1/CADCHF

1/AUD USD = USD CAD

Nice stuff R. I like the visual comparison. It gives far more info than a single number.

Is anyone trading BTCUSD and any TRY pairs? Trends look extremely strong in those. I don't see any long-term trends, bullish or bearish, in any other instrument.

dax, dow and its friends?

yes of course the stock indexes, great bull trends!

my point was, I don't know how people trade the usual gold/oil/major fx pairs... except for GBPUSD (bearish), none of them have a clear trend on the longer time-frames. If the Monthly/Weekly time-frame is choppy, then even if you spot a trend on for e.g. 4H or even Daily, the odds of winning go down drastically.

BTW, anyone know who provides the best spreads for TRY? From what I've seen the spreads are at least 15 ticks on different brokers, which is insane.

I wish I was. Thanks for the tip. :-)

I am closely looking at Guido's (Sesamstrasse) suggestions though.

Dow is firing on all cylinders though with a nice recent momentum burst.....



I have my eyes on Nikkei.....might have a play soon.



GBPUSD is starting to shape up and may be a candidate in the next few hours.



Nothing else is sticking out on my watchlist.

GU is still on the playlist demo and live . Gcad closed too early --

and GN broke the 195 level .. attacking the BRN?

Nice Guido

This is for you mate :-)

Hi C

Could you please tell me the equation how to calculate different percentages of these 2 systems if we would like to achieve x drawdown? Of course according to this set of data. It can change in future, but to have approx. idea.

For example if I put 0.19% risk for reverter system, most of the time it will be higher, because of my small capital and minimum 0.01 lot.
So if minimum position is 0.01 lot and this creates let's say 0.5% of risk then EDTT should have risk of (0.5/0.19)*1.31=3.44% for short term and long term SDC? Then the possible drawdown would be 26.31%? Or it is more complex to calculate?

Also it is difficult since there is no SL in reverter system...

Hi R

Yep mate...but it takes a bit of explanation so I will step through a comprehensive example tomorrow night. It's probably a good time to explain how to do it on this thread and then we will have a permanent record of it here if anyone else wants to do this. Better still I will run through an example of how to set up your spreadsheet to derive some useful statistics that can help in your portfolio construction.

Stay tuned mate :-)

PS Just getting into a bit of early Xmas spirit tonight mate with the kids putting up a christmas tree while trying to multi-task watching the Secret Life of Walter Mitty at the same time. Man.....I love that movie :-)

Some note from Andreas' book Stocks on the move:

Just read the financial news during a busy day in the market for a demonstration. It can be quite comical at times really, especially when there’s some more significant news coming out. At first, the market might be at -0.2% in the morning. News ticker updates with “Market down on FED fears”. Two hours later, the market is at +0.2% and the news ticker proclaims “Market positive on FED hopes”. The announcements finally come out, and the market dips down to -0.5%. News tells us that the market is falling on FED disappointment. As the market finally closes at +0.5% the news concludes that the market rallied on the positive FED announcement.

Hehehe. The news is always right.......eventually :-)

"Blending of Return Streams - How to Do it"

.....continued from post 2501.....Part 1



I thought it may be a good time to have a post or two that looked at the fairly simple methods that are used to blend the return streams of a diverse array of assets to compile powerful portfolios solutions.

The first thing to understand is that a return stream simply represents the sequence of profits of losses generated by a single asset or multi asset that is ordered in a time series. What is critical however to ensure that all return streams are treated equally is to apply percentage returns as opposed to $ returns.

To organise your trade data into a time series to analyse the performance of a composite of return streams, the first thing you have to do is as follows:
1.Sort your trade data into separate return streams in a time ordered basis;
2 Determine the frequency of compounding you would like to do under a live portfolio. For example, I apply compounding principles to my portfolios only on a monthly basis. Some people may want to compound more frequency.

Once you have determined your desired compounding frequency then you need to summate the $ returns of each distinct return series for each month and then calculate the percentage return of that month in relation to your trade capital.

For example let's say I have 1000 trades in a time series using 24 different assets for a fixed trade capital of $200,000, I would separate those $ returns into separate returns streams and then consolidated the trades into monthly $ results for each return stream and then calculate the $Return/$Trade Capital for each separate month.

For example 5 trades in January 2017 occurred with say EURUSD of which a total profit/loss for the month was $2,500. My monthly return for EURUSD for January was therefore 2500/200000 = 1.25%.

Now when you test a portolio it is wise to adopt a standard trade risk percentage in your testing simply to generate return streams of a know trade risk.....and then later you can multiply these returns of dilute those returns through different position sizing.

For example below is a matrix of trade results that has compiled trade results into $ returns for a number of instruments and a number of months.



Once you have it ordered in $ per month per series with an understanding of your trade capital, you can then convert it into percentage returns.



Now remember how we tested this portfolio with a defined trade risk of 0.25% or $500/$200,000 which was used to derive our position sizing per trade. If we wanted to doubled the returns of the series, we would simply use a 0.50% trade risk and the % results of that return series will double. Having an understanding of your trade risk % and the distribution of returns you can then derive different weights for each return series through position sizing.

I might leave it here for a while as I have a touch of the tummy wobbles and need to get some shut eye.

Stay tuned.

C

I had great Friday today (Flynn/Trump news).



I noticed, that if I would use 3/4 ADR buy/sell stops instead of 1/2 ADR over 500 Donchian, most of these losses would be avoided. I cannot avoid all whipsaws, but these were just ridiculous to the point almost....

Like this baby:

My problem was not US news but CAD behaviour.
Killed all my relative trades!

As I said two days ago, some of these spikes will eventually be our way...