Price movements in the same direction?

[oxusmorouz]

Nope. That one will catch the reversal days as well. Try this one instead

Today's close > Yesterday's close AND Today's close >=0.35*(H-L) and <= 0.65*(H-L)

Regards,
Kalyan.

Alas, if only I had the data for open-high-low from 1897-1970. I have OHLC data only from 1970 onwards.

 

[oxusmorouz]

hi,

what if we only consider

yesterday's close > close of day before and today's close > yesterday's close

~Winston

Ahh sorry Winstonn, I have badly misunderstood what you meant.
In this case, what do I use as the base?

 

[oxusmorouz]

I think the probability would increase with

yesterday's close < close of day before and today's close > yesterday's close

Regards,
Kalyan.

1 - .52 = .48 (as already stated) would be the probability in the case where price moves in the opposite direction the next day.

Had the given probability far different (below or above) the 50% figure, I guess we can come to some conclusion of the "cause-effect" relationship (If I am right) which exists between price movements of 2 successive days.

Alas, from the given result (without hypothesis testing), I am inclinded to say that one day's price movement has no connection with the next day's price movement since the probability of both (for and against) tends to 50%, generated from an unbiased coin toss.

 

[CreditViolet]

Oh my, whats going on here

 

[kkseal]

Today's close > Yesterday's close AND Today's close >=0.35*(H-L) and <= 0.65*(H-L)

Ahh, i got it a bit wrong here This one is for gauzing the next day's(tomorrow's) price direction; but the way you are plotting it the condition would be

(Yesterday's close > Day Before yesterday's close) AND (Yesterday's close >=0.35*(H-L) and <= 0.65*(H-L))

Yeah you did say you did not have the OHLC data for the test period, but why not test it for 1970 onwards?

Likewise modifying the other one to

day before yesterday's close < close of previous day and yesterday's close > day before yesterday's close

Its unidirectional (UP) however.

Regards,
Kalyan.

 

[kkseal]

Oh my, whats going on here

Mumbo Jumbo testing

 

[kenneth]

Nothing Much CV they are just expressing their view of the pobability of the price direction..... proabably they are confusing you...and ... Me

Is there a easy solution ? an AFL in your collection?

Regards
Ken

 

[kkseal]

Nothing Much CV they are just expressing their view of the pobability of the price direction..... proabably they are confusing you...and ... Me

I there a easy solution ? an AFL in your collection?

Regards
Ken

Don't think CV gets confused that easily

An AFL will be easy enough to code Let's see if there's anything in this first.
Also it's unidirectional as of now.

Regards,
Kalyan.

 

[kenneth]

Will this help?

Regards
Ken

 

[CreditViolet]

Don't think CV gets confused that easily

An AFL will be easy enough to code Let's see if there's anything in this first.
Also it's unidirectional as of now.

Regards,
Kalyan.

I think its a gud idea overall, however if one is simply interested in price persistence, a simple runs test will do. I dont think one can find any meaningful/significant probability relationships with the above-mentioned approach.How do I know? Been there, done that, got the trophy

Other concepts like gaps might be more meaningful.