Where is the line between fitting and actual patterns? - page 41

 
joo:

Again:


Thanks again!

joo:

...... on the abscissa axis is the ordinal number of the occurrence of the pattern,


If I understand correctly, maybe it is better like this ....... on the abscissa axis is the order number of the observation of the combination of events (i.e. the phenomenon of the pattern).

.......

In the figure - the occurrence of the pattern, I was confused at first.

 

No, we should leave it as it is.

 
joo:

Hence, the relationships of the phenomena (one phenomenon is followed by the corresponding one) will look like this:

_a->a_,

_b->b_,

_c->c_,

..........

_j->j_,

But, the market is a very complex non-stationary thing, in which there are rarely one hundred percent correspondences of phenomena, so, a qualitative characteristic k of a particular pattern would be something like:

k(Q:A)=(_a->a_)*100%.

If you say.

Then another question, now on the ordinate axis:

If _a->a_ is an observable pair of events, then we can only talk about binary outcomes {0;1}.

How does the characteristic k form such a distribution?

 
lasso:

If you say so.

Then another question, now on the ordinate axis:

if _a->a_ is an observable pair of events, then we can only talk about binary outcomes {0;1}.

How does the characteristic k form such a distribution?

"Sit down, Vovochka. Two! Reread your synopsis, you're confusing bears with camels again.":)

 

All right, teacher! :-)

I'll follow the outline strictly:

======================

Event _a.1 = the fast fMA has crossed the slow tMA downwards

Event _a.2 = the difference of the kotir value and the value of the crossing point _a.1 > 40 points

Event _a.3 = Present time < 12:00

The combination (conjunction) of events _a.1 and _a.2 and _a.3 is the phenomenon _a of the pattern A

--------->

Event a_.1 = the difference of the quote value at the moment of the event _a.1 and the quote value at the moment of the event a_.1 > 50 pips (i.e., sell position in the profit)

Event a_.2 = Current time = 21:00 (second type)

The combination (conjunction) of events a_.1 and a_.2 is the phenomenon of the a_ pattern A

==========

So, on 21.01.2011(dt0), we observed the phenomenon _a(dt0) and a_(dt0) and calculated that k(Q:A)=(_a(dt0)->a_(dt0))*100% = 100%

============

And on 26.01.2011(dt1), we observed the _a(dt1) phenomenon, BUT at 21:00 the price did not reach the level of the a_.1(dt1) event by some 25 points.

...............................................................

The scavenger! Well, how do you find the percentage of correspondence between two related phenomena on 26.01.2011?

================

P/S/ I sent my presumptive answer to sever30, for self-checking... ))

 
lasso:

Ok, teacher! :-)

I will strictly follow the outline:

======================

Event _a.1 = Fast fMA has crossed the slow tMA downwards

Event _a.2 = the difference between the value of the quotient and the value of the intersection point _a.1 > 40 points

Event _a.3 =Current time < 12:00

The combination (union) of events _a.1 and _a.2 and _a.3 is the phenomenon _a of the pattern A

--------->

Event a_.1 = the difference of the quote value at the moment of the event _a.1 and the quote value at the moment of the event a_.1 > 50 pips (i.e., sell position in the profit)

Event a_.2 = Current time = 21:00 (second type)

The combination (conjunction) of events a_.1 and a_.2 is the phenomenon of the a_ pattern A

==========

So, on 21.01.2011(dt0), we observed the phenomenon _a(dt0) and a_(dt0) and calculated that k(Q:A)=(_a(dt0)->a_(dt0))*100% = 100%

============

And on 26.01.2011(dt1), we observed the _a(dt1) phenomenon, BUT at 21:00 the price did not reach the level of the a_.1(dt1) event by some 25 points.

...............................................................

The scavenger! Well, how do you find the percentage of correspondence between two related phenomena on 26.01.2011?

================

P/S/ I sent my presumptive answer to sever30in my personal email, for self-checking... ))


You've clogged up the thread.
 
lasso:

P/S/ I sent my presumptive answer to sever30in a private message, for self-checking... ))


:)

the north has no idea what you're talking about here :)

 
sever30:

:)

The north has no idea what you're talking about here :)

You don't have to.

Just let you keep.... Don't delete it. )

 
paukas:
You've clogged up the thread.

What's this?

An attempt to get back into the thread? ;- )

Yeah, come on in... We welcome everyone.... ))

..........................................

"Come in don't be afraid. Leave don't cry."

 
lasso:

OK, teacher! :-)

I will strictly follow the outline:

======================

Event _a.1 = fast fMA has crossed slow tMA downwards

Event _a.2 = the difference between the value of the quotient and the value of the intersection point _a.1 > 40 points

Event _a.3 =Current time < 12:00

The combination (union) of events _a.1 and _a.2 and _a.3 is the phenomenon _a of the pattern A

--------->

Event a_.1 = the difference of the quote value at the moment of the event _a.1 and the quote value at the moment of the event a_.1 > 50 pips (i.e., sell position in the profit)

Event a_.2 = Current time = 21:00 (second type)

The combination (conjunction) of events a_.1 and a_.2 is the phenomenon of the a_ pattern A

==========

So, on 21.01.2011(dt0), we observed the phenomenon _a(dt0) and a_(dt0) and calculated that k(Q:A)=(_a(dt0)->a_(dt0))*100% = 100%

============

And on 26.01.2011(dt1), we observed the _a(dt1) phenomenon, BUT at 21:00 the price did not reach the level of the a_.1(dt1) event by some 25 points.

...............................................................

The scavenger! Well, how do you find the percentage of correspondence between two related phenomena on 26.01.2011?

================

P/S/ I sent my presumptive answer to sever30in my personal email, for self-checking... ))

If it is possible to estimate the degree of correspondence of one phenomenon to another (it depends on what set of events as a phenomenon is chosen in TS), by correlation for example, then it is possible to express this degree of correspondence in % ratio (such version I presented). If it is not possible (either there is a response phenomenon, or not) - then it remains only to count, how many correspondences on Sample.

The CU master architect is the lord, whatever is convenient.

PS. And hey... what the hell kind of teacher am I?