Numerical series density - page 7

New comment
 
-Aleks-:

Got the following numbers(though there's still a zero in there - we'll have to figure out the reason why):

The reason is understood - everything is correct - zero from the fact that there is a gap in the groups.

NO.P./P.NumberDeltaClose valuesClose in successionMaximumDenseDensityDensity v2
1 10 4 10
2 13 3 1 1 13
3 15 2 1 2 15
4 21 6 1 3 21 2,20 2,75
5 40 19 0 0 40
6 42 2 1 1 42
7 46 4 1 2 46
8 51 5 1 3 51
9 56 5 1 4 56 3,20 4,00
10 71 15 0 0 0
11 78 7 0 0 78
12 81 3 1 1 81 0,60 0,75

As you can see from the table, the largest density of 0.75 (0.6) is in the smallest group of two numbers, 78 and 81, and the largest group with a density of 4 (3.2) from numbers 40 to 56.

 

When processing real data, you may have a dilemma - what to do with the same numbers. I think it makes more sense to change them by an insignificant amount of hundredths, but you have to check how many of these numbers will go in a row...

 

I ask the moderator to move the topic to theAutomated Trading Systems section - this topic was created to solve a specific problem, directly related to the construction of the ATS.

 
-Aleks-:

I ask the moderator to move the topic to"Automated Trading Systems" - this topic was created to solve a specific problem directly related to the construction of the ATS.

It is possible to find density in a simple way, if there are no RETURNED numbers.

1. find the average

2. we look where there are more numbers, on the right or on the left of the average

3. i.e. we have found basic density. Suppose there are more numbers on the left. Now we find the average of them.

4. We get 2 more areas, where there are more numbers, and they are denser. Total 4 areas we have found already in order of density

 

If you just want to find numbers that are close to each other, then put them in order of increasing number, calculate the deltas one by one

and simply by the difference of the deltas we mark, delta 5 is one colour, delta 3 is another colour, delta 1 is even denser.

 
Vyacheslav Kornev:

It is possible to find the density in a simple way if there are no RETURNED numbers.

1. find the average

2. we look where there are more numbers, on the right or on the left of the average.

3. i.e. we have found basic density. Suppose there are more numbers on the left. Now we find the average of them.

4. We get 2 more areas, where there are more numbers, and they are denser. Total 4 areas we have found already in order of density

Try your method at least in Excel on the numbers given here and share the result, please.

 
-Aleks-:

Try your method at least in Excel on the numbers given here and share the result, please.

OK, I'll do it later, probably not until next weekend
 

In the meantime, I suggest another method for finding the maximum density.

We need to calculate the distance from the number of all the other numbers, that is, we have a row of 12 numbers, so

For the first number count all 11 deltas, all distances to other 11 numbers. Add up all its deltas.

For the second, we add all its 11 deltas, and then we add ALL its deltas.

for the third and so on.

And that number/numbers which have the least total delta are the densest

I did it with 4 numbers.

 

more accurately you will not find the tightest points of any row

only way to find the point that is closest to all the others - i.e. where the density is.

Am I good?)

 
1234567891011121314...26
New comment