SQX 116 - Monte Carlo - option to Filter results on either 'Individual' or 'Composite' MC results

Hi Mark

Currently with SQX, filtering of MC condition values, when selecting 'multiple' MC simulation types, filters based on the 'composite' result values taken from all selected MV simulations together. Is it possible to add selection feature that would allow filtering of results to be based on both:
1. Results from each 'Individual' MC simulation; also include feature for user to view MC plot for each individual simulation
2. Results from 'Composite' MC simulations (as currently implemented)

Thanks
AC1962

Attachments
No attachments
  • Votes +6
  • Project StrategyQuant X
  • Type Feature
  • Status Archived
  • Priority Normal
  • Assignee None

History

AC
#1

AC1962

23.12.2018 15:15

Task created

h
#2

hankeys

23.12.2018 16:53
Voted for this task.
KL
#3

kainc301

20.01.2019 11:00
Voted for this task.
KL
#4

kainc301

20.01.2019 11:01
Unified analysis is a great idea
Rr
#5

Partizanas

20.01.2019 11:42
Voted for this task.
MF
#6

Mark Fric

02.05.2019 15:40

Milestone changed from Build 121 to Build 122

MF
#7

Mark Fric

14.06.2019 10:35

Milestone changed from Build 122 to None

AC
#8

AC1962

29.12.2019 11:00
Hi Mark

Please advise Milestone for this feature request? Milestone was previously 122, but is currently None.

Thanks

AC1962

JK
#9

Insanity82007

02.01.2020 08:30
Voted for this task.
JK
#10

Insanity82007

02.01.2020 13:45
This would be a really great option for both MC crosschecks because the results seem to compound when using more than 1 option under each MC cross check menu.


For example, under MC retests, if I select randomise params, randomise history, randomise spread and randomise my MC results for 200 iterations appear to be compounded / composite and I will get anything from -3x Ret/DD to +2x Ret/DD on my main strategy and this makes it impossible to set a reliable filter because the level of excessive randomness will never happen in real life.

Instead I would like all 4 of these MC tests to be individually assessed.


This falls in line with what I'm trying to achieve by asking for: https://roadmap.strategyquant.com/tasks/sq4_5831

JK
#11

Insanity82007

02.01.2020 13:50

Attachment MC Retest Example.png added

For example my best and worst MC restest results from 200 simulations with the 4 MC tests I mentioned above (image attached)
KP
#12

CooleRnax

08.01.2020 09:42
Voted for this task.
MF
#13

Mark Fric

26.11.2020 10:50

Status changed from New to Archived

AC
#14

AC1962

27.11.2020 00:51
Hi Mark

What do you actually mean by Archived?

Is my suggestion still being considered for future, or has it been shelved?

Thanks

AC1962

b
#15

bentra

27.11.2020 05:45
Voted for this task.

Votes: +6

Drop files to upload

or

choose files

Max size: 5MB

Not allowed: exe, msi, application, reg, php, js, htaccess, htpasswd, gitignore

...
Wait please