pystatsmodels
https://groups.google.com/d/forum/pystatsmodels
We are discussing the implementation of statistical and econometrics models and methods of data handling in Python using statsmodels.enRe: [pystatsmodels] Bug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/N8YnOciQDQAJ
Perfect. Thanks so much On Friday, September 30, 2016, <josef...@gmail.com> wrote: > On Fri, Sep 30, 2016 at 8:25 PM, Stuart Reynolds > <stuart@stuartreynolds.net> wrote: > > Here's a good (Bayesian) way of think about it: > > > > > http://stats.stackexchange.com/questions/13225/what-is-the-distrhttps://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
Stuart ReynoldsSat, 01 Oct 2016 05:52:51 UTCRe: [pystatsmodels] Bug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/7hONQuuADQAJ
On Fri, Sep 30, 2016 at 8:25 PM, Stuart Reynolds <stu...@stuartreynolds.net> wrote: > Here's a good (Bayesian) way of think about it: > > http://stats.stackexchange.com/questions/13225/what-is-the-distribution-of-the-binomial-distribution-parameter-p-given-a-samp > http://www.sumsar.net/blohttps://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
josefpktdSat, 01 Oct 2016 01:02:09 UTCRe: [pystatsmodels] Bug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/Znux6-Z-DQAJ
Here's a good (Bayesian) way of think about it: http://stats.stackexchange.com/questions/13225/what-is-the-distribution-of-the-binomial-distribution-parameter-p-given-a-samp http://www.sumsar.net/blog/2014/01/bayesian-first-aid-binomial-test/ (which referenence to R package) It (I thinkhttps://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
Stuart ReynoldsSat, 01 Oct 2016 00:25:12 UTCRe: [pystatsmodels] Bug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/bY_32V3sDAAJ
On Wed, Sep 28, 2016 at 10:59 PM, Stuart Reynolds <stuart....@gmail.com> wrote: > Thanks for the reply. There's a typo in my original post. > > "Shouldn't it be the case that Pr(p>x) == 1-Pr(p>x)?" > ==> > "Shouldn't it be the case that Pr(p>x) == 1-Pr(p<x)?" > > I think you understood, buthttps://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
josefpktdThu, 29 Sep 2016 03:39:54 UTCRe: [pystatsmodels] Bug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/c6019i7qDAAJ
Thanks for the reply. There's a typo in my original post. "Shouldn't it be the case that Pr(p>x) == 1-Pr(p>x)?" ==> "Shouldn't it be the case that Pr(p>x) == 1-Pr(p<x)?" I think you understood, but just to be super clear, the answers aren't symmetric as the must be. If: x: 0.1 Pr(p<x)=0.001,https://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
Stuart ReynoldsThu, 29 Sep 2016 02:59:54 UTCRe: [pystatsmodels] Bug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/4Zr-Tt7nDAAJ
binomial >> variable is >x (actually, I'd prefer to know Pr( x_lower > p && p < x_upper >> ) ). >> >> I think that statsmodels.stats.proportion.binom_test calculates this, but >> here's its output: >> >> k= 3 (observed successes) >> n= 3 (nobs) >> x: 0.0 Pr(p<x) 0.0 Pr(p>x), 1.0 >> x:https://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
josefpktdThu, 29 Sep 2016 02:17:28 UTCRe: [pystatsmodels] Bug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/uWKInjPmDAAJ
On Wed, Sep 28, 2016 at 7:45 PM, Stuart Reynolds <stuart....@gmail.com> wrote: > I'm trying to test the likelihood that probability of success for a binomial > variable is >x (actually, I'd prefer to know Pr( x_lower > p && p < x_upper > ) ). > > I think that statsmodels.stats.proportion.bihttps://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
josefpktdThu, 29 Sep 2016 01:46:56 UTCBug in statsmodels.stats.proportion.binom_test? (or brain fart)
https://groups.google.com/d/msg/pystatsmodels/QaQc88bX718/u_AER1HgDAAJ
I'm trying to test the likelihood that probability of success for a binomial variable is >x (actually, I'd prefer to know Pr( x_lower > p && p < x_upper ) ). I think that statsmodels.stats.proportion.binom_test calculates this, but here's its output: k= 3 (observed successes) n= 3 (nobs) x:https://groups.google.com/d/topic/pystatsmodels/QaQc88bX718
Stuart ReynoldsWed, 28 Sep 2016 23:59:06 UTCRe: [pystatsmodels] ImportError: No module named 'statsmodels.tsa.regime_switching._hamilton_filter'
https://groups.google.com/d/msg/pystatsmodels/Uy3v96WrdYA/iyDDoE9eBwAJ
Thanks a lot! That helped :) Best, Aleks > > and needs cython and a c compiler > > Josef >https://groups.google.com/d/topic/pystatsmodels/Uy3v96WrdYA
Aleksandar KarakasTue, 27 Sep 2016 13:10:39 UTCRe: [pystatsmodels] ImportError: No module named 'statsmodels.tsa.regime_switching._hamilton_filter'
https://groups.google.com/d/msg/pystatsmodels/Uy3v96WrdYA/2azkGRMqBwAJ
On Mon, Sep 26, 2016 at 4:40 PM, Aleksandar Karakas <aleks....@gmail.com> wrote: > Hello! > > After the latest rebase of my branch [1] I am getting an exception [2]. > According to GitHub my branch is up to date with statsmodels/master and > statsmodels/master is passing all tests on GitHub.https://groups.google.com/d/topic/pystatsmodels/Uy3v96WrdYA
josefpktdMon, 26 Sep 2016 21:13:24 UTCImportError: No module named 'statsmodels.tsa.regime_switching._hamilton_filter'
https://groups.google.com/d/msg/pystatsmodels/Uy3v96WrdYA/96s-PUYoBwAJ
Hello! After the latest rebase of my branch [1] I am getting an exception [2]. According to GitHub my branch is up to date with statsmodels/master and statsmodels/master is passing all tests on GitHub. Does anyone know where the problem is coming from? Aleks [1] pushed to https://github.com/yoghttps://groups.google.com/d/topic/pystatsmodels/Uy3v96WrdYA
Aleksandar KarakasMon, 26 Sep 2016 20:40:25 UTCRe: [pystatsmodels] Statsmodels VAR(p) : Clarification on how it works
https://groups.google.com/d/msg/pystatsmodels/qNKVaO2ODUo/z_av8yY8BgAJ
On Fri, Sep 23, 2016 at 2:58 PM, paw <patricia....@gmail.com> wrote: > Hi, > > I am trying to use Statsmodels VAR(p) function, but I just want to make > sure on how it works. > > I am new to this so bear with me. > > The general equation of VAR(p): > > > VAR(p): Yt = A1Yt-1+ …+ApYt-p + noisehttps://groups.google.com/d/topic/pystatsmodels/qNKVaO2ODUo
josefpktdFri, 23 Sep 2016 20:33:26 UTCStatsmodels VAR(p) : Clarification on how it works
https://groups.google.com/d/msg/pystatsmodels/qNKVaO2ODUo/HAfFKTw7BgAJ
Hi, I am trying to use Statsmodels VAR(p) function, but I just want to make sure on how it works. I am new to this so bear with me. The general equation of VAR(p): VAR(p): Yt = A1Yt-1+ …+ApYt-p + noise From what I understand, if I have data that looks like below and p = 3 (just as anhttps://groups.google.com/d/topic/pystatsmodels/qNKVaO2ODUo
pawFri, 23 Sep 2016 20:16:37 UTCRe: [pystatsmodels] Discrepancy of Negative Binomial coefficient
https://groups.google.com/d/msg/pystatsmodels/MSbVlDwduME/uJYnIOgmBQAJ
Hi, moreover, there is still a hugh difference in the change rate, computed by the mean and formula for the change rate ( (new value - old value)/old value) Simple way in Excel: from 2 to 3= 0.3639 from 2 to 4= 2.5568 NB regression: from 2 to 3= 0.4529 from 2 to 4= 1.2385 Any explanation forhttps://groups.google.com/d/topic/pystatsmodels/MSbVlDwduME
tronTue, 20 Sep 2016 07:52:51 UTCRe: [pystatsmodels] Discrepancy of Negative Binomial coefficient
https://groups.google.com/d/msg/pystatsmodels/MSbVlDwduME/682QQbjbBAAJ
Hi, Am Sonntag, 18. September 2016 18:31:08 UTC+2 schrieb josefpktd: > > On Sun, Sep 18, 2016 at 10:38 AM, 'tron' via pystatsmodels > <pystat...@googlegroups.com> wrote: > > Hi, > > > > can I really use the coefficients by using this categorical version as > > change rates? Or do I have tohttps://groups.google.com/d/topic/pystatsmodels/MSbVlDwduME
tronMon, 19 Sep 2016 08:55:02 UTC