ASX Stock Pairs Trade Journal

[sleepy]

Hi skc/walt,

What sort of returns have you been getting recently?
And are you finding pairs trading as profitable as when you first started?

sleepy

 

[Pairs Trader]

New trade
Long AMP @ 4.87
Short PPT @ 30.60
.

Sold AMP @ 5.15 (5.74% gain)
Covered PPT @ 30.28 (1.05% gain)

 

[skc]

Hi skc/walt,

What sort of returns have you been getting recently?
And are you finding pairs trading as profitable as when you first started?

sleepy

Pretty much business as usual at my end. Profitability (adjusted for positoin size) is similar to when I first started.

 

[FXanyone]

Hi skc/walt,

What sort of returns have you been getting recently?
And are you finding pairs trading as profitable as when you first started?

sleepy

I have been doing about as well. Frankly my improvements are coming from better control of losses. I have traded for many years but have found the setting of pair stops to be tricky.

 

[edman79]

How are you controlling your losses FX?

 

[Pairs Trader]

Hi skc/walt,

What sort of returns have you been getting recently?
And are you finding pairs trading as profitable as when you first started?

sleepy

Whilst a significant portion of your profits can be attributed to market conditions, the majority of your profits come from you, the trader, and your ability to consistently execute a trading plan day in, day out and get better at your craft.

 

[Pairs Trader]

I have been doing about as well. Frankly my improvements are coming from better control of losses. I have traded for many years but have found the setting of pair stops to be tricky.

How are you settings stops? Ive always found $ and % stops to decrease pair trading performance like most trading systems, however a time stop can mitigate those large losses whilst not damaging performance.

 

[FXanyone]

How are you controlling your losses FX?

Finding the most effective stop is on individual legs rather than the pair. Also if not seeing start of a reversion with a week I cut.

 

[sleepy]

Finding the most effective stop is on individual legs rather than the pair.

Has anyone done any testing on the merits of closing both trades at same time (via say revision to mean, or a time stop) vs using a trailing stop on the one(s) in profit and only exiting when taken out.

sleepy

 

[SilverRanger]

Has anyone done any testing on the merits of closing both trades at same time (via say revision to mean, or a time stop) vs using a trailing stop on the one(s) in profit and only exiting when taken out.

sleepy

It might be a pain to do this manually, given most brokers (as far as I'm aware) don't provide much support for pair trading.

With IB I used to be able to trade in pair (buy and sell as a pair, as opposed to individual legs), and I think you can put a trailing stop there. IB has now removed this function "due to increased volatility in the market".

I suppose you may be able to do this with a broker offering DDE in Excel, where you can basically define your pair trailing stop with some fancy formulas.

Closed my first and only pair opened in August MND/MRM today with 4% profit (vs July with 7%)

 

[skc]

Closed a trade that I opened on Monday. Long FGL 6.06, closed 6.38. Short GFF 1.395, covered 1.40.

Actually I am holding half of the Fosters - the bid for the wine business is probably just the starting gun for some serious bidding on the beer business. But the trade is closed for record keeping purpose at a solid 5% over 2 days.

When I can long a stock that is on the M&A radar, I pretty much always jump to it.

Another trade I am holding is long WHC / short AQA. Someone wanted to buy WHC - hopefully in the next few days now that the election is finally over.

 

[FXanyone]

Closed a trade that I opened on Monday. Long FGL 6.06, closed 6.38. Short GFF 1.395, covered 1.40.

Actually I am holding half of the Fosters - the bid for the wine business is probably just the starting gun for some serious bidding on the beer business. But the trade is closed for record keeping purpose at a solid 5% over 2 days.

When I can long a stock that is on the M&A radar, I pretty much always jump to it.

Another trade I am holding is long WHC / short AQA. Someone wanted to buy WHC - hopefully in the next few days now that the election is finally over.

I'm having a reasonable start to the month. Only one horror ( MGX/FMG) and some solid little trades. Currently have QAN/MAP STO/OSH ORI/IPL (long/short)

 

[sleepy]

hi skc,

What universe of stocks did you look at when setting up your pairs?
e.g., ASX 200 - GCIS sectors

... or did you use GCIS Industries?

And how many pairs do you currently have on your watchlist?

sleepy

 

[skc]

I'm having a reasonable start to the month. Only one horror ( MGX/FMG) and some solid little trades. Currently have QAN/MAP STO/OSH ORI/IPL (long/short)

I hope you got out of STO in time before the 5% plunge. I actuall had the same trade (STO/OSH) and got out pretty much breakeven.

Funny reaction but I suppose the price was not as good as expected, or that it means STO is another step closer to the inevitable capital raising?

hi skc,

What universe of stocks did you look at when setting up your pairs?
e.g., ASX 200 - GCIS sectors

... or did you use GCIS Industries?

And how many pairs do you currently have on your watchlist?

sleepy

I use GCIS as a guide but I go a little bit deeper with my pairing. E.g. I don't just group all mining stocks, but only ones that mine the same material. And capital goods are not all the same etc etc.

I have ~300 pairs in the program, but many I don't trade even if the signal comes up. I use the program as a market monitor to alert me what stocks are on the move relative to its peers.

 

[sleepy]

Thanks skc,

Do you stay within ASX 200only?
Or are you more flexible?
And do you flush and rinse all your pairs say every 2-3 months?

And how many signals on average would you get a day?
And of these how many would you take vs. disregard after checking news, ratio charts etc

Just trying to get a handle on what sort numbers I should be expecting if I had 200-300 pairs on watch.

sleepy

 

[skc]

Thanks skc,

Do you stay within ASX 200only?
Or are you more flexible?
And do you flush and rinse all your pairs say every 2-3 months?

And how many signals on average would you get a day?
And of these how many would you take vs. disregard after checking news, ratio charts etc

Just trying to get a handle on what sort numbers I should be expecting if I had 200-300 pairs on watch.

sleepy

I very rarely go beyond ASX200.

I tend to get 15-20 signals per day. Many of these will be repeats. E.g. STO slumped today so all the other oil and gas stocks it paired against will be triggered. I then check the ratio charts to see which pair has the most divergence, the most suitable news environment, and the most suitable looking price chart. (Note I didn't take the STO trade today, although it will probably bounce a bit tomorrow).

On average I open 2.4 trades per day, and close 2.2 trades. The max was 8 new trades in a day. I hold on average 9.2 pairs, with max of 16 pairs open at any one time - that was prob too many.

 

[sleepy]

Hi skc,

Thanks for the helpful info. I think I need to rethink the way I set my pairs as Im not getting that amount of signals with ASX200 stocks (backtested 5yrs).

At the moment the only filters Im using are:

avg profit > $300
correlation > 0.70
sd > +/- 2.70

E.g. STO slumped today so all the other oil and gas stocks it paired against will be triggered. I then check the ratio charts to see which pair has the most divergence, the most suitable news environment, and the most suitable looking price chart. (Note I didn't take the STO trade today, although it will probably bounce a bit tomorrow).

For example, I have just done 5yr backtest (after market close) on 'S&P ASX 200 SubInd Oil & Gas Exploration & Production' stocks which comprises:

AWE
BPT
CVN
ESG
KAR
LNC
OSH
ROC
STO
WPL

...and only one pair containing STO .... OSH/STO looked like it could have possiblly triggered today. It is currently: (correlation = 0.80, sd = +2.45).

sleepy

 

[sleepy]

Hi skc,

Looks like I found the problem.
Changing SD to +- 2.00 produces a bunch more signals (e.g., 4 STO pairs today).

Would you recommend using SD of +- 2.00 or sticking with the default setting of
+- 2.70?

And do you ONLY take signals that show divergence on or is this not a must have.

sleepy

 

[FXanyone]

I hope you got out of STO in time before the 5% plunge. I actuall had the same trade (STO/OSH) and got out pretty much breakeven.

Funny reaction but I suppose the price was not as good as expected, or that it means STO is another step closer to the inevitable capital raising?



I use GCIS as a guide but I go a little bit deeper with my pairing. E.g. I don't just group all mining stocks, but only ones that mine the same material. And capital goods are not all the same etc etc.

I have ~300 pairs in the program, but many I don't trade even if the signal comes up. I use the program as a market monitor to alert me what stocks are on the move relative to its peers.

Got out of STO as it went back through the previous days close. I guess if analysts value the stake at $ 1 billion + and you sell for $ 650 mill the asset is worth less and you have to have a bigger capital raising

 

[SilverRanger]

Picked up my first pair of the month at market close today. Long DJS @ 5.06, short MYR @ 3.92.