Lets all join the Millionaires CLub
- Thread starter FetteredChinos
- Start date
- Watchers 2
ColinRiche
Well-known member
- Messages
- 284
- Likes
- 3
Correction to my previous post
There are 4 gaps in that dow data in the first post
17/9/01
5/3/99
10/9/98
8/4/96
.
There are 4 gaps in that dow data in the first post
17/9/01
5/3/99
10/9/98
8/4/96
.
ColinRiche
Well-known member
- Messages
- 284
- Likes
- 3
JonnyT said:
JonnyT,
To help you visualise a days data simply as possible
OHLC data can be represented by 3 lines for each day
With an \/\ or /\/ shape
For an up day draw a line from the OPEN down to the LOD up to the HOD and then down to the CLOSE.
For a down day draw a line from the OPEN up to the HOD down to the LOD and up to the CLOSE
As you can see for a DOWN day the HOD comes first
and for an UP day the LOD comes first
Hope this helps
.
Last edited:
Theirs a flaw in my code!!!
The result of 780 points was flawed because I used the HOD/LOD to the point in time rather than the high/low of each candle/
When this bug was corrected the result is only 400 odd points.
I have tried StopLosses offsets, all sorts really and I cannot get a tradeable system from this. I beleive the method to be next to useless.
I have attached my spreadsheet complete with code.
JonnyT
The result of 780 points was flawed because I used the HOD/LOD to the point in time rather than the high/low of each candle/
When this bug was corrected the result is only 400 odd points.
I have tried StopLosses offsets, all sorts really and I cannot get a tradeable system from this. I beleive the method to be next to useless.
I have attached my spreadsheet complete with code.
JonnyT
FetteredChinos
Veteren member
- Messages
- 3,897
- Likes
- 40
cheers jonny, i concede on this one. it appears that 2003 hasnt been the best year for the system... trending for 10 months hasnt helped.. i'll look into this further, perhaps adding some further filters. thanks for ure help though.
FC
FC
FetteredChinos
Veteren member
- Messages
- 3,897
- Likes
- 40
vagaries of data feeds
ok guys, i know this has been done to death before, but im sure there is something seriously weird going on here...
i decided to do this visually to save time.
have a look at this link
http://custom.marketwatch.com/custo...ize=4&type=2&lf2=0&lf3=0&comp1=&comp2=&comp3=
if that link doesnt work, im attaching a gif from the FT website.
the point id like to highlight is the fact that look how many times the close of the day is in between the range of the previous day. i think i counted 40 out of 55 days where this happened when the previous range had been exceeded..
i thought it was generally held that FT had a decent data source.
the picture on the charts of the DJIA seems to back up my initial testing. 40 from 55 days is 73%, which is roughly in line with what my backtesting from yahoo seemed to show, and the losses where it didnt close between the original range, dont APPEAR to be that high...
this is all visually, of course.. Wheelbarrow of salt required etc etc.
something is very odd here. The only way to really know is by forward testing. Im gonna give it 20 trades or so to see if it is a) tradable, and b) worth my while doing so!
could be an expensive mistake, but hey ho, its only money. moer important things in life... (such as that ****ing speedboat)
looks like im gonna have to go on Bullseye. bring back Jim Bowen
all is forgiven
FC
ok guys, i know this has been done to death before, but im sure there is something seriously weird going on here...
i decided to do this visually to save time.
have a look at this link
http://custom.marketwatch.com/custo...ize=4&type=2&lf2=0&lf3=0&comp1=&comp2=&comp3=
if that link doesnt work, im attaching a gif from the FT website.
the point id like to highlight is the fact that look how many times the close of the day is in between the range of the previous day. i think i counted 40 out of 55 days where this happened when the previous range had been exceeded..
i thought it was generally held that FT had a decent data source.
the picture on the charts of the DJIA seems to back up my initial testing. 40 from 55 days is 73%, which is roughly in line with what my backtesting from yahoo seemed to show, and the losses where it didnt close between the original range, dont APPEAR to be that high...
this is all visually, of course.. Wheelbarrow of salt required etc etc.
something is very odd here. The only way to really know is by forward testing. Im gonna give it 20 trades or so to see if it is a) tradable, and b) worth my while doing so!
could be an expensive mistake, but hey ho, its only money. moer important things in life... (such as that ****ing speedboat)
looks like im gonna have to go on Bullseye. bring back Jim Bowen
all is forgivenFC
You will find that the FT charts for intraday are correct and for EOD they are not,
The EOD charts seem to include futures data.
This has been the case for a long time (2 years that I know of)
You have to use futures data as that is what you trade and not the cash index.
www.downloadquotes.com has highly accurate EOD data for the cash indices, the others are flawed.
But remeber you cannot trade the cash indices...
Hey you want to scrape around then trade this system but it sure wont make you a million.
JonnyT
The EOD charts seem to include futures data.
This has been the case for a long time (2 years that I know of)
You have to use futures data as that is what you trade and not the cash index.
www.downloadquotes.com has highly accurate EOD data for the cash indices, the others are flawed.
But remeber you cannot trade the cash indices...
Hey you want to scrape around then trade this system but it sure wont make you a million.
JonnyT
FetteredChinos
Veteren member
- Messages
- 3,897
- Likes
- 40
ok JT thanks...
ure quite right about this system. a million curses on yahoo!
just when i thught i was getting somewhere....
but still, if FT are using futures data, there is still a chance of being triggered out of hours...
i have a cunning plan batman!!
at least this thinking is keeping me off the streets..
oops, im late for a meeting.
FC
ure quite right about this system. a million curses on yahoo!
just when i thught i was getting somewhere....
but still, if FT are using futures data, there is still a chance of being triggered out of hours...
i have a cunning plan batman!!
at least this thinking is keeping me off the streets..
oops, im late for a meeting.
FC
TheBramble
Legendary member
- Messages
- 8,394
- Likes
- 1,170
JonnyT said:
You can SB them.
ColinRiche
Well-known member
- Messages
- 284
- Likes
- 3
There not cash prices there futures prices that the SB companies use Bramble
.
.
TheBramble
Legendary member
- Messages
- 8,394
- Likes
- 1,170
My mistake then Colin.
I was looking at D4F's screen and they 'apparently' offer the "US30 Index Cash".
Presumably this is just 'their version' of the current value.
I've also just had a look at IG's 'value' - I see what you mean.
As an aside, has/does/can anyone play the differentials between the various SBs for the same instrument?
I can't think what strategy might work, but IG seem somewhat more clued up than D4F (today, on the DJ anyway) in that IG will switch to the high end of the bid/offer when the Dow's moving up and back to the low end when its falling. Very sensible from their point of view. But the interesting thing is that D4F (apart from having a much smaller spread (5pts as opposed to IG's 9pts) move their quote far more slowly.
Just a thought. I'd be very interested in any strategies that might take advantage of this - be they hypothetical, academic or worked!
I was looking at D4F's screen and they 'apparently' offer the "US30 Index Cash".
Presumably this is just 'their version' of the current value.
I've also just had a look at IG's 'value' - I see what you mean.
As an aside, has/does/can anyone play the differentials between the various SBs for the same instrument?
I can't think what strategy might work, but IG seem somewhat more clued up than D4F (today, on the DJ anyway) in that IG will switch to the high end of the bid/offer when the Dow's moving up and back to the low end when its falling. Very sensible from their point of view. But the interesting thing is that D4F (apart from having a much smaller spread (5pts as opposed to IG's 9pts) move their quote far more slowly.
Just a thought. I'd be very interested in any strategies that might take advantage of this - be they hypothetical, academic or worked!
Bluewave
Active member
- Messages
- 240
- Likes
- 2
Yeah FC, looks like ur quids in today. Shame I didn't get in on the drop though I always seem to be looking at the wrong market at the wrong time!
Today was a classic case of the fundamentals ruling over the technicals....this trading game is a bit hard isn't it?
I always seem to be looking at the wrong market at the wrong time!Today was a classic case of the fundamentals ruling over the technicals....this trading game is a bit hard isn't it?
ColinRiche
Well-known member
- Messages
- 284
- Likes
- 3
Bramble
Arbitrage of the prices from different spreadbetting companies was discussed at length elsewhere on these boards
It was concluded that this was not a viable option and far too risky
.
Arbitrage of the prices from different spreadbetting companies was discussed at length elsewhere on these boards
It was concluded that this was not a viable option and far too risky
.
ColinRiche
Well-known member
- Messages
- 284
- Likes
- 3
Now keeping this thread on topic....
I have a dow chart yesterday (Tuesday) and EOD values from Sharescope
O 10701.1 H 10748.8 L 10579.3 C 10609.9
Which i know aren't cash values - could be futures values...i dont know
So what values do we have from elsewhere to compare ?
.
I have a dow chart yesterday (Tuesday) and EOD values from Sharescope
O 10701.1 H 10748.8 L 10579.3 C 10609.9
Which i know aren't cash values - could be futures values...i dont know
So what values do we have from elsewhere to compare ?
.
ColinRiche
Well-known member
- Messages
- 284
- Likes
- 3
rglenn said:
FC's idea is that it makes money if it closes within yesterday's range.
It would of been the Long that was triggered and so today this system would have lost money
.
Bigbusiness
Experienced member
- Messages
- 1,408
- Likes
- 23
I seem to repeat this a lot so it should probably go in a FAQ section.
In the good old days, the high and low of the Dow was worked out using the high and low of the constituent stocks. I think yahoo, bigcharts etc. use this method after the market has closed, so that today's data can be compared to the data from the early 1900's.
As not all stocks trade at the high or low of the day at the same time, there is a discrepancy between the historical data and the index data seen on an intra-day chart.
It's not futures data as if you look closely, every candlestick has a top and a tail. Sometimes the futures gap down after the close and never go above the previous days close price, so it can't be futures data.
Hope that makes sense. I don't know for sure if this is correct but it makes more sense than any other explanation I have seen.
To illustrate this, on the left is my excel chart using OHLC data. On the right is the same section using the other type of data. It makes a huge difference to the candlesticks.
In the good old days, the high and low of the Dow was worked out using the high and low of the constituent stocks. I think yahoo, bigcharts etc. use this method after the market has closed, so that today's data can be compared to the data from the early 1900's.
As not all stocks trade at the high or low of the day at the same time, there is a discrepancy between the historical data and the index data seen on an intra-day chart.
It's not futures data as if you look closely, every candlestick has a top and a tail. Sometimes the futures gap down after the close and never go above the previous days close price, so it can't be futures data.
Hope that makes sense. I don't know for sure if this is correct but it makes more sense than any other explanation I have seen.
To illustrate this, on the left is my excel chart using OHLC data. On the right is the same section using the other type of data. It makes a huge difference to the candlesticks.
Attachments
Last edited:
Big Business
you are right about the 'restatement' of H/L
and the Open as well because they dont all start trading at the same time. Several may have not started trading at 2.45.
but it is not at all clear which site uses it or not.
It has always seemed a 'purists' approach to me because
the Index did not actually reach those points.
It is something of an illusion,imo
And you certainly cant trade them.
well, maybe some guy in Hicksville ,Pa makes a book ?
you are right about the 'restatement' of H/L
and the Open as well because they dont all start trading at the same time. Several may have not started trading at 2.45.
but it is not at all clear which site uses it or not.
It has always seemed a 'purists' approach to me because
the Index did not actually reach those points.
It is something of an illusion,imo
And you certainly cant trade them.
well, maybe some guy in Hicksville ,Pa makes a book ?
Similar threads
