The Day Of Week effect, does it still work

[dtrader98]

Quote from MGJ:

Yes, 6/1/1982 is a Tuesday. It is also the next market day after Friday, 5/28/1982. That's what Joe wrote and that's what I tested. Joe entered at the open on Tuesday 6/1/1982 and exited at the close, and so did my spreadsheet. You can see this for yourself by looking at the other page within the Excel file (excerpted below).



"Since I used Excel's WEEKDAY function to identify Fridays, then selected the market-days-after-Friday to put in the worksheet named (drum roll.....) "The_Day_After_Friday", a missing Monday doesn't throw off the entire rest of the spreadsheet.

By the way, Murray was wrong when he said that "Monday is a holiday" is the only problem with Joe's algorithm. For ten points, name one other problem. For five thousand points, name two other problems.

Aaahh .. yes. I did misunderstand your original premise. For some reason, I was under the assumption that the thread was about trading on a specific "day of the week" (o-cl) only (from earlier posts), and whether monday would be the best compounded trading day of the week.
I should have read the details of your backtest on joe's strategy more carefully Although, it does seem kind of odd that he would call it "buy monday" when it includes non-mondays. Would make more sense to call it, buy the next market day following friday IMO.



The graph I posted shows the result of trading each specific day of the week, under the assumptions of slippage and spread gain you presented.

From my recollection of l. williams, I have seen him specifically address best trading days of the week and edge advantage. "Joe's" algorithm, as you presented it, slightly deviates from that premise, but nevertheless, some good points are raised about it from other posters. One thing I noticed off the bat about running the specific days of the week, is that Wednesday shows better long term compounded results, but the data is skewed by the advantage of many more trading Wednesdays in that time period, and thus Wednesday's results gain from the advantage of additional compounded days (although, one could argue it could have been adversely effected by a large cluster of negative days).

 

[Gyles]

Very informative thread.

Just wondering whether it is possible to do "The Day of the Week"on the "Thirty Year Treasury"?

Please do clarify.

 

[ForexGuru]

Another losing day for this "Day Of Week" strategy, that's 2 losing days out of 2 since this post was started. Maybe better luck this Friday and Monday.

How about posting paper trades every Friday and Monday for 52 weeks then we can see how it works over a 1 year period?

 

[dtrader98]

Hopefully, I'm not overloading on image size here, but it was about the best i could fit on one page.

Here's one year's running view of data for S&P 500 to present, 2007-- the way I approach it. Have at it. Double check it.
All refunds available on proof of error

<img src="http://elitetrader.com/vb/attachment.php?s=&postid=1583307" border="0" alt=""><br /></font></p></font></p></font></p></font></p></font></p>

 

[Murray Ruggiero]

Quote from dtrader98:

Hopefully, I'm not overloading on image size here, but it was about the best i could fit on one page.

Here's all the data you want to see for S&P 500 to present, 2007, the way I approach it. Have at it.

Thank's , it looks like you did a nice job.

 

[NihabaAshi]

Quote from ForexGuru:

Another losing day for this "Day Of Week" strategy, that's 2 losing days out of 2 since this post was started. Maybe better luck this Friday and Monday.

How about posting paper trades every Friday and Monday for 52 weeks then we can see how it works over a 1 year period?

dtrader98 stats had already shown that Wednesday is the day to play.

http://www.elitetrader.com/vb/showthread.php?s=&threadid=102124&perpage=5&pagenumber=6

Also, we are talking about a tendency from Open to Close and such may not be practical as is.

I myself look at like this...

Instead of playing the the Open to Close, look for bullish signals anytime during the trading day along with being more weighted on the Long positions in comparison to the Short position.

Simply, I tend to use the market seasonal patterns for position size managment along with holding the trades much longer in comparison to the opposite trades.

Thus, on Wednesdays I would be larger on my Long positions in comparison to my Short positions.

In addition, I would try to capture more profits on my Long positions in comparison to my Short positions along with having a tendency to add to profitable Long positions in comparison to not doing any adding to profitable Short positions.

Therefore, whatever is your primary trade signal...

Try to merge it with the market seasonal pattern.

It's much easier to manage your trading in comparison to try to manage a tendency that may look good in the stats but not practical for real trading as is.

See prior discussion of a trade management result of another discussed market seasonal pattern to better understand what I meant by merging a trade signal with the market seasonal pattern.

http://www.elitetrader.com/vb/showthread.php?s=&postid=1564663#post1564663

Thus, some market seasonal patterns are not meant (not practical for real trading conditions) to be traded as is while others are ok for such.

I believe (won't know for sure until I complete my research) this tendency for Wednesday's to be an up day is one of those tendencies not to trade as is and requires merging with an actual bullish trade signal to know when to go Long.

Mark

 

[ForexGuru]

Quote from NihabaAshi:

dtrader98 stats had already shown that Wednesday is the day to play.

From post #3:

"In my research on day of week effect in the S&P500 which I originally did back in 1995, I found that Monday had an upward bias and Friday downward bias..... The Monday ,Friday bias has a lot to do with the fear since the crash of 1987, that is when these relationships really got strong. People don't want to hold over the weekend and they buy positions back on Monday. "

Lets not hop from day to day.

 

[NihabaAshi]

Quote from ForexGuru:

From post #3:

Lets not hop from day to day.

I specifically said dtrader98 stats in reference to his Wednesday's numbers that's in contrast with what your quoting Murray.

So far...Wednesday merits my research for trading purposes in comparison to what the stats show about the other trading days of the week.

Quote from NihabaAshi:

I'm strongly interested into what key market events occurred in late 2001 to change things for Wednesday.

I'm also curious about 1998 key market events that changed things for Tuesday.

Knowing such can be useful.

Mark

I also have a secondary interest in Tuesday's statistics but not for trading purposes.

Other than that, I haven't seen any reasons to start my research with Monday's and Friday's.

Mark

Both charts below originally posted by dtrader98 in this thread.

<img src="http://elitetrader.com/vb/attachment.php?s=&postid=1580669" border="0" alt=""><br /></font></p></font></p></font></p></font></p>

<img src="http://elitetrader.com/vb/attachment.php?s=&postid=1583307" border="0" alt=""><br /></font></p></font></p></font></p></font></p></font></p>

 

[rcanfiel]

Quote from ForexGuru:

Another losing day for this "Day Of Week" strategy, that's 2 losing days out of 2 since this post was started. Maybe better luck this Friday and Monday.

How about posting paper trades every Friday and Monday for 52 weeks then we can see how it works over a 1 year period?

Why are you trading this? Things like this are rarely going to pour money into your coffers. And especially when in the midst of outsized volatility.

You should only use this in concert with other methods. It is interesting, not a trading methodology

 

[ForexGuru]

Quote from rcanfiel:

Why are you trading this? Things like this are rarely going to pour money into your coffers. And especially when in the midst of outsized volatility.

You should only use this in concert with other methods. It is interesting, not a trading methodology

I am not trading the "Day Of Week" strategy. I am 1 vote of 31 for No it's a myth.