The Day Of Week effect, does it still work

Murray - are you abandoning this thread? I am interested in this type of analysis - I had previously tested most of what has been shown so far but I am interested where you might take the analysis. I'm always on the lookout for new ideas.

I see there are a few flamers of these seasonal style ideas but I think there is some validity personally. The Monday effect appears to have died - it has not shown any significant edge for many years (including the slight variation of buy monday if Friday was an up day which was a stunning performer for a while). The Friday effect itself still shows some edge as a negative day but, as Mark was saying, is not strong enough to be traded "as is".

The reason i liked the Friday / Monday effect was that it was based on some logic - ie fear of holding over the weekend. The "Wednesday" effect which has been mooted on this thread is interesting but until I hear a logical hypothesis of why exists, I would be inclined to put it down as a "random" effect in the data and certainly would not trade it.

 

The sole purpose of starting this thread was to get people interested in TradersStudio. The Day Of Week effect is - and was always - a myth.

 

This thread was so interesting, what has happened? Is anyone still interested? Have noticed that the “Wednesday”, day of this week was “big down day” (low day) for trading, as quoted below:

Source: The Press Association: Home sales data sends Dow down

What is your opinion regarding this?

 

Indeed it was a down day. The statistics show, that 2 out of 3 Wednesday's are up days (67%). Better odds than any other day of the week for the current year. Possibly, the holiday shift had some influence -- who knows?
Either way, the last two Wednesdays were up. By virtue of the odds, it makes total sense to have a down Wednesday.

Too many posters are reading this thread and not understanding basic probability and statistics. You can't expect every single Wednesday to be up when the odds say 2/3 are up. In order for those odds to be valid, every 1/3 must be down.
And again, that doesn't apply sequentially either (that is a conditional probability study you need to undertake, if you want more granularity).

However, when posters say it is a myth, it clearly shows they don't understand the study in front of their eyes, nor how to interpret it.
Anyone can take a basic statistics course, or read a book and run the study for themselves. As I mentioned, it is not subjective, meaning if someone else runs it, they should get the exact same results, and if they don't they can come to the board and dispute the results with their counter study.

Not hollow words, or thoughts, but an actual concrete and verifiable study.

 

Good luck in fantasy land.

 

Either way, the last two Wednesdays were up. By virtue of the odds, it makes total sense to have a down Wednesday.

It is just sad...

 

Actually, what's sad is when trolls come to threads and begin lambasting studies, with a tone of authority, while offering absolutely zero evidence towards rebuttal.

Suppose someone inquisitively asks, what's the sum of two plus two? You take the time to demonstrate that 2+2 =4, proofs and all. The inquisitor thinks on it a while, then comes back and says, you are an absolute charlatan, I personally added another 2 and it's no longer equal to 4.

What do you say to that? The counter argument is not even grounded in the same reference plane.

If anyone want to contribute any meaningful ideas that can be discussed with a modicum of rational logic (cynics included), feel free to continue the thread. Otherwise, no reply necessary.

 

Quote from dtrader98:

"Actually, what's sad is when trolls come to threads and begin lambasting studies, with a tone of authority, while offering absolutely zero evidence towards rebuttal."

Your grasp of statistics is weak. That is what is sad.

If you post, and are semi-clueless, you are going to attract negative posts. Call them trolls, but there are also people who post uneducated drivel, and expect to be taken seriously.

 

"Evidence" from us "trolls" to lambasting "studies" is not needed. The market will provided the "evidence" that "The Day Of Week effect" is in fact a myth and it has done that since this thread was started almost 1 month ago.

 

This was another active topic when I was updating it should this be addressed in the future.