Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

3 Pages123>
Options
Go to last post Go to first unread
Offline Davide Carpi  
#1 Posted : 05 November 2012 14:30:15(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
Hi all,

Just a seed for statistical purposes Happy


Plugin handles primitives listed below:

Sample Estimators

Probability Density Functions

Cumulative Density Functions

Quantile functions

Random Numbers

Tools



PLEASE REPORT HERE ANY ISSUE.


requirements: .Net Framework 3.5 (Windows) / Mono 2.0 (Linux) / SMath Studio 0.97.5581
installation: Tools > Plugins > change "Local Storage" to "Online Gallery" > Statistical Tools
sources: if you want to see the plugin sources, look in the SVN repository.
embedded plugins: Meta Numerics library 2.2.1.0 (website, license)

Edited by user 15 September 2015 16:31:42(UTC)  | Reason: Meta Numerics

File Attachment(s):
StatisticalTools_docs.zip (35kb) downloaded 304 time(s).
Davide Carpi attached the following image(s):
CDF.gif
If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 05/11/2012(UTC)
Offline omorr  
#2 Posted : 05 November 2012 15:01:07(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,735
Man
Serbia

Was thanked: 313 time(s) in 267 post(s)
Hello w3b5urf3r,

I remained almost speechless by seeing this Dance Biggrinancing: Biggrinrinks: Shok

Have no words to express my gratitude for this plugin Thank you

Just hope that the inverse cumulative distributions would be applied as well.

Just one suggestion if you would not mind. There is a naming convention in Mathcad and R ("lingua franca" for statistics).

pdf (probability density starts with d - like dnorm)
cdf (cumulative probability function starts with p - like pnorm)
inverse cdf (inverse cumulative probability function starts with q - like qnorm)
random numbers from a distribution starting with r - like rnorm

Regards,
Radovan

EDIT: Maybe redundant to say, but besides normal there are t-student, chi-squared and F which are "must have" continuous distributions.

Edited by user 05 November 2012 16:02:17(UTC)  | Reason: Not specified

When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
thanks 1 user thanked omorr for this useful post.
on 06/11/2012(UTC)
Offline mkraska  
#3 Posted : 05 November 2012 22:08:53(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,648
Germany

Was thanked: 915 time(s) in 575 post(s)
Hello w3b5urf3r,

thank you for the great work. I just got one question. As far as I understand, the Mode function gives the value (or a vector of values) with the highest number of occurences in a dataset.

Then, in the following the result should be a vector of (1 2 3 5) with 4 occurences.

Mode(mat(1,3,5,1,6,2,3,2,5,2,3,5,5,2,1,3,4,1,3,6),1)=mat(mat(5,1,1),mat(2,1,1),2,1)@#

Best regards, Martin

BTW, I like the smart handling of vector arguments in functions!

Edited by user 05 November 2012 22:12:09(UTC)  | Reason: Not specified

Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
Offline Davide Carpi  
#4 Posted : 06 November 2012 02:17:40(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
Originally Posted by: mkraska Go to Quoted Post
Hello w3b5urf3r,

thank you for the great work. I just got one question. As far as I understand, the Mode function gives the value (or a vector of values) with the highest number of occurences in a dataset.

Then, in the following the result should be a vector of (1 2 3 5) with 4 occurences.

Mode(mat(1,3,5,1,6,2,3,2,5,2,3,5,5,2,1,3,4,1,3,6),1)=mat(mat(5,1,1),mat(2,1,1),2,1)@#

Best regards, Martin

BTW, I like the smart handling of vector arguments in functions!


Thank you Martin, I've lost 2 lines of code cleaning the sources...

fixed & updated Good


best regards

w3b5urf3r
If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
Offline mkraska  
#5 Posted : 07 November 2012 02:23:55(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,648
Germany

Was thanked: 915 time(s) in 575 post(s)
Hello w3b5urf3r,

some minor doc issues:

in pdf.sm, the doc for pdf.uniform needs a little correction:

f(x,a,Cool:line(if(a>b,line(swap:a,a:b,b:swap,3,1),0),if((x≥a)&(x≤Cool,1/{b-a},0),2,1)

Also, the pdf.Rayleigh(2) and cdf.Rayleigh(2) help strings seem to have a copy-paste error.

Best regards, Martin

Edited by user 07 November 2012 18:45:38(UTC)  | Reason: Not specified

Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
Offline omorr  
#6 Posted : 07 November 2012 12:03:45(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,735
Man
Serbia

Was thanked: 313 time(s) in 267 post(s)
Hello w3b5urf3r,

I would kindly ask you to insert into this plugin at least intercept(x,y) and slope(x,y) functions for calculating the least square intercept and slope of a straight line giving two data vectors.

Regards,
Radovan

Edited by user 07 November 2012 12:05:52(UTC)  | Reason: Not specified

When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
Offline Davide Carpi  
#7 Posted : 08 November 2012 17:36:35(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
Originally Posted by: omorr Go to Quoted Post
Hello w3b5urf3r,

I would kindly ask you to insert into this plugin at least intercept(x,y) and slope(x,y) functions for calculating the least square intercept and slope of a straight line giving two data vectors.

Regards,
Radovan

Ok Biggrin .

Originally Posted by: mkraska Go to Quoted Post
Hello w3b5urf3r,

some minor doc issues:

in pdf.sm, the doc for pdf.uniform needs a little correction:

f(x,a,Cool:line(if(a>b,line(swap:a,a:b,b:swap,3,1),0),if((x≥a)&(x≤Cool,1/{b-a},0),2,1)

Also, the pdf.Rayleigh(2) and cdf.Rayleigh(2) help strings seem to have a copy-paste error.

Best regards, Martin

Done, will be available in the next relase Good .


regards,

w3b5urf3r

Edited by user 08 November 2012 19:12:21(UTC)  | Reason: Not specified

Davide Carpi attached the following image(s):
SMath Studio - [sample estimators.sm].png
If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 08/11/2012(UTC)
Offline Davide Carpi  
#8 Posted : 02 December 2012 15:07:42(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
PLUGIN UPDATED

- added more tools: Regularized Beta function, Regularized Gamma functions (P and Q), Ceil, Floor, error function, inverse error function;
- added more estimators: Kurtosis, Kurtosis Excess, Moment, Skewness, Intercept, Slope;
- added more probability distributions: Student's t, Fisher-Snedecor, Chi-squared, Cauchy;
- added quantiles (all distributions except Student's t, Fisher-Snedecor and Chi-squared).


best regards,

w3b5urf3r

Edited by user 02 December 2012 16:48:06(UTC)  | Reason: Not specified

If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 02/12/2012(UTC)
Offline omorr  
#9 Posted : 02 December 2012 21:43:33(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,735
Man
Serbia

Was thanked: 313 time(s) in 267 post(s)
Hello w3b5urf3r,

Many thanks for this plugin Good .

Just played a bit, and have seen a problem with CDF.ChiSquare( "variable" )-(image attached). Did not quite understand why this problem appear at the moment. Also, do not quite understand what this description mean

CDF.ChiSquare( "variable" ) - Returns the Χ² sDOF CDF evaluated in "1:variable" points.

I suppose that sDOF means "single degree of freedom"

Regards,
Radovan

Edited by user 02 December 2012 21:45:16(UTC)  | Reason: Not specified

omorr attached the following image(s):
cumulative density functions.png
When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
thanks 1 user thanked omorr for this useful post.
on 08/12/2012(UTC)
Offline Davide Carpi  
#10 Posted : 02 December 2012 23:02:26(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
Originally Posted by: omorr Go to Quoted Post
Hello w3b5urf3r,

Many thanks for this plugin Good .

Just played a bit, and have seen a problem with CDF.ChiSquare( "variable" )-(image attached). Did not quite understand why this problem appear at the moment. Also, do not quite understand what this description mean

CDF.ChiSquare( "variable" ) - Returns the Χ² sDOF CDF evaluated in "1:variable" points.

I suppose that sDOF means "single degree of freedom"

Regards,
Radovan


Biggrin seem that some eval() in the Reguarized Gamma functions it's stricly necessary for some value.... fixed and updated Wink

You supposed right Tongue Now I've extended the acronym Wink

EDIT: plugin updated again (Mean and StdDev names)

regards,

w3b5urf3r

Edited by user 03 December 2012 02:53:14(UTC)  | Reason: Not specified

Davide Carpi attached the following image(s):
StatisticalTools plugin - Samples - SMath Studio Forum - Mozilla Firefox.png
SMath Studio - [Page1].png
If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 03/12/2012(UTC)
Offline omorr  
#11 Posted : 03 December 2012 11:21:12(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,735
Man
Serbia

Was thanked: 313 time(s) in 267 post(s)
Hello w3b5urf3r,

I wonder why all of the functions inside plugin are not in a separate function category. They are inside "Matrix and vector" function category now.
Not so important at the moment, I am just curious Good

Regards,
Radovan

P.S. I am looking forward to the missing ICDF's Good Thank you
When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
Offline Davide Carpi  
#12 Posted : 03 December 2012 13:14:54(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
Originally Posted by: omorr Go to Quoted Post
Hello w3b5urf3r,

I wonder why all of the functions inside plugin are not in a separate function category. They are inside "Matrix and vector" function category now.
Not so important at the moment, I am just curious Good

Regards,
Radovan

P.S. I am looking forward to the missing ICDF's Good Thank you


Hi omorr,

I'd like to put them in a new category "Statistics" but I do not know if it is possible to expand the available categories through the plugin. Because all functions accept a vector as input i've grouped all as "matrix and vector".

About the missing ICDF I'm looking for a fast and accurate algorithm to compute each of them... if I find anything I will use a bisection algorithm Good


regards,

w3b5urf3r
If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 03/12/2012(UTC)
Offline turducken  
#13 Posted : 07 February 2013 22:16:18(UTC)
turducken

Rank: Newbie

Groups: Registered
Joined: 07/02/2013(UTC)
Posts: 4

Will this plugin eventually have the ability to produce a random number from a normal or Gaussian distribution (rather than uniform distribution)?


I would be using this to create random white Gaussian noise to simulate measured/actual situations. For instance, Mathcad has the function rnorm() that does this.

Thank you, and I greatly appreciate SMath and am looking forward to transitioning to it once I understand it has all the capabilities I need.
Offline Davide Carpi  
#14 Posted : 08 February 2013 13:50:45(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
Originally Posted by: turducken Go to Quoted Post
Will this plugin eventually have the ability to produce a random number from a normal or Gaussian distribution (rather than uniform distribution)?

I would be using this to create random white Gaussian noise to simulate measured/actual situations. For instance, Mathcad has the function rnorm() that does this.

Thank you, and I greatly appreciate SMath and am looking forward to transitioning to it once I understand it has all the capabilities I need.


Sure, soon or later... Good


best regards,

w3b5urf3r
If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
Offline Davide Carpi  
#15 Posted : 04 June 2013 09:51:08(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
A Nightly update for anyone are using SMath Studio 0.96.4902 (plugins -> online gallery) and for anyone are using the previous SMath Studio relases (1st post).

- improvements of existing Random() functions (seed issues)
- Added a Guid (after the download an error should appear using worksheets created with the old plugin, just continue and save again the file)
- Sources are moved to the SVN server (there is a link in the first post)


best regards,

w3b5urf3r

Edited by user 04 June 2013 13:44:06(UTC)  | Reason: Not specified

If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 04/06/2013(UTC)
Offline mkraska  
#16 Posted : 03 September 2013 16:20:39(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,648
Germany

Was thanked: 915 time(s) in 575 post(s)
There is a problem with plotting the Gamma function, see attachment. Might be an issue of plotregion rather than Statistical tools...

Edited by user 03 September 2013 16:23:36(UTC)  | Reason: Not specified

File Attachment(s):
Gamma.sm (9kb) downloaded 52 time(s).
mkraska attached the following image(s):
Gamma.PNG
Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
Offline Davide Carpi  
#17 Posted : 03 September 2013 17:13:12(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
Gamma(1) is a built-in function, don't know what's the issue No
If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 03/09/2013(UTC)
Offline mkraska  
#18 Posted : 03 September 2013 18:59:58(UTC)
mkraska


Rank: Advanced Member

Groups: Registered
Joined: 15/04/2012(UTC)
Posts: 1,648
Germany

Was thanked: 915 time(s) in 575 post(s)
Originally Posted by: w3b5urf3r_reloaded Go to Quoted Post
Gamma(1) is a built-in function, don't know what's the issue No

Sorry, I was tricked by the capital letter...
One more reason to mention the plugin in the description string in order to avoid such confusion. Sometimes I check plugin origin of a given function by switching off the guessed one and trying if I still get the function. As that is not really handy, I tend to be lazy just like in the given case :-(

Martin Kraska

Pre-configured portable distribution of SMath Studio: https://en.smath.info/wi...th%20with%20Plugins.ashx
Offline Davide Carpi  
#19 Posted : 15 September 2015 11:52:13(UTC)
Davide Carpi


Rank: Advanced Member

Groups: Registered, Advanced Member
Joined: 13/01/2012(UTC)
Posts: 2,322
Man
Italy
Location: Italy

Was thanked: 1167 time(s) in 768 post(s)
PLUGIN UPDATED SMath Studio 0.97.5581

  • .NET Framework 3.5 required;
  • (SS-91) Added missing quantiles (Chi Square, Student's t, F) using the Meta Numerics library 2.2.1.0 (website, license);
  • erf, erfc, erfinv from Mata.Numerics;
  • minor code changes;

Edited by user 15 September 2015 11:57:06(UTC)  | Reason: Not specified

If you like my plugins please consider a donation to SMath Studio; for personal contributions to me: paypal.me/dcprojects
thanks 1 user thanked Davide Carpi for this useful post.
on 15/09/2015(UTC)
Offline omorr  
#20 Posted : 15 September 2015 19:32:05(UTC)
omorr


Rank: Administration

Groups: Registered, Administrators, Advanced Member
Joined: 23/06/2009(UTC)
Posts: 1,735
Man
Serbia

Was thanked: 313 time(s) in 267 post(s)
Thank you Davide,

At last...long awaited quantile functions Good

Best Regards,
Radovan
When Sisyphus climbed to the top of a hill, they said: "Wrong boulder!"
thanks 1 user thanked omorr for this useful post.
on 16/09/2015(UTC)
Users browsing this topic
3 Pages123>
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.