Hi!
Thank you for your coments.
>
> > Support symbolic expressions of density function that implement the
> covariance matrix.
> Can you provide some examples for this idea.
>
If we have a normal multinomial random variable X = (X1, X2), with
covariance matrix Σ and mean μ, a symbolic expression of the density
function that implements the covariance matrix will be: 2pi *
(det(Σ))^(-1/2) * exp((-1/2)*(X-μ)'*Σ*(X-μ)) (For the generic case that X
can have arbitrary length the form i was thinking is here
<https://wikimedia.org/api/rest_v1/media/math/render/svg/c66e6f6abd66698181e114a4b00da97446efd3c4>).
My reasoning for this is that i have found this kind of expressions more
useful and easy to understand, no matter how large the covariance matrix
is.
>
>
> > I also noticed that in the joint_rv_types for some classes there are
> corresponding functions that return a JointRandomSymbol object with a call
> to the multivariate_rv function buts that's not the case for other classes.
> Is there a reason for it or it is something that has not yet been
> implemented?
> It would be easy to comment if you can give some examples.
>
One example I found is in the joint_rv_types file, where the
NormalGammaDistribution has a corresponding function (not as part of the
class) called NormalGamma but for the MultivariateLaplaceDistribution
there’s no such function.
One more thing that i wanted to ask you was about the "Support integration
and equation solving for common probability tasks" section in the ideas
page. Should we find the cases, where the cdf can’t be found by sympy, even
thought it exists, and hard code in the solution or try to re-write the
code in a way that sympy can find the answer?
I will soon post a compelet draft of my proposal.
Thank you very much for the help and the feedback!
>
> On Fri, Mar 20, 2020 at 6:03 PM Basilis Kalos <[email protected]
> <javascript:>> wrote:
>
>> Hi!
>> I am working on my application and i was thinking of ways to extend the
>> support of multivariate distributions. For example:
>> 1. Add more matrices support.
>> a. Use matrices modules to perform some basic operations for
>> the multivariate rv's, for example we could use the eigen module to
>> perform
>> Principal Component Analysis and Factor Analysis.
>> b. Add functions to calculate covariance and correlation
>> matrices.
>> c. Support symbolic expressions of density function that
>> implement the covariance matrix.
>> 2. Add sampling for multivariate distribution by using and
>> extending upon the numpy libraries.
>> Does this sound good?
>>
>> I also noticed that in the joint_rv_types for some classes there are
>> corresponding functions that return a JointRandomSymbol object with a call
>> to the multivariate_rv function buts that's not the case for other classes.
>> Is there a reason for it or it is something that has not yet been
>> implemented?
>>
>> Thank you in advance!
>>
>> Τη Τρίτη, 17 Μαρτίου 2020 - 8:31:07 π.μ. UTC+2, ο χρήστης Gagandeep Singh
>> (B17CS021) έγραψε:
>>>
>>> Hi,
>>>
>>> As far as I know, the "Probability" project is available for GSoC, 2020.
>>> Please let me know of any questions regarding GSoC projects related to
>>> `stats` module.
>>>
>>> Best wishes.
>>>
>>> On Tue, Mar 17, 2020 at 1:06 AM Basilis Kalos <[email protected]>
>>> wrote:
>>>
>>>> Hi all
>>>>
>>>>
>>>> The first project that i’m most interested to work on is the
>>>> “Optimize floating point expressions”. I am familiar with Herbie and I
>>>> have
>>>> started reading its source code with the purpose of adapting ideas and
>>>> even
>>>> code (after re-writing it in python). How does that sound?
>>>>
>>>>
>>>> I also would be really excited to work on the “Probability” project.
>>>> Are any of those project ideas available? Should i look for projects that
>>>> no-one else is working on?
>>>>
>>>>
>>>> Thank you!
>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "sympy" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send
>>>> an email to [email protected].
>>>> To view this discussion on the web visit
>>>> https://groups.google.com/d/msgid/sympy/0e580f98-f87e-4183-b078-b1fdaa2a2376%40googlegroups.com
>>>>
>>>> <https://groups.google.com/d/msgid/sympy/0e580f98-f87e-4183-b078-b1fdaa2a2376%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>> .
>>>>
>>>
>>>
>>> --
>>> With regards,
>>> Gagandeep Singh
>>> Github - https://github.com/czgdp1807/
>>> Linkedin - https://www.linkedin.com/in/czgdp1807/
>>> <https://www.linkedin.com/in/gdp1/>
>>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "sympy" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected] <javascript:>.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/sympy/af2c649b-d13e-45fe-a1b5-a657131da877%40googlegroups.com
>>
>> <https://groups.google.com/d/msgid/sympy/af2c649b-d13e-45fe-a1b5-a657131da877%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>
>
> --
> With regards,
> Gagandeep Singh
> Github - https://github.com/czgdp1807/
> Linkedin - https://www.linkedin.com/in/czgdp1807/
> <https://www.linkedin.com/in/gdp1/>
>
--
You received this message because you are subscribed to the Google Groups
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/1144b7c0-7370-415b-8fdb-fcc2ade7e912%40googlegroups.com.
