Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: list coefficient estimations
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: list coefficient estimations
Date
Thu, 14 Apr 2011 15:31:12 +0100
I think this desire is statistically ridiculous, frankly, but just
looking at the help for -matrix list- would show that you can tune the
display. You could also use hexadecimal format to see every last bit
that Stata is holding for you.
. sysuse auto
(1978 Automobile Data)
. gen gpm = 1 / mpg
. regress gpm weight
. mat li e(b)
e(b)[1,2]
weight _cons
y1 .00001407 .00770767
. help matrix
. mat li e(b), format(%23.18f)
e(b)[1,2]
weight _cons
y1 0.000014070444284905 0.00770766777955807
. mat li e(b), format(%21x)
e(b)[1,2]
weight _cons
y1 +1.d8203232cddfdX-011 +1.f921350aa8958X-008
On Thu, Apr 14, 2011 at 3:23 PM, Max Fotbollen <[email protected]> wrote:
> This is what I get for my coefficients;
> mat li e(b)
> e(b)[1,2]
> Storks _cons
> y1 4.4352163 1401.127
> Which is by the way the same nember of decimal points that you get in the output directly after running the model.
> Corresponing figure from SPSS storks' coefficient is 4.434216270161429.
>
>
>
> And here is what I ANOVA figures, which are ok.
>
> ereturn list
> scalars:
> e(N) = 16
> e(df_m) = 1
> e(df_r) = 14
> e(F) = 226.840002418306
> e(r2) = .9418701218260083
> e(rmse) = 415.5066042516718
> e(mss) = 39162959.66552542
> e(rss) = 2417040.334474576
> e(r2_a) = .9377179876707231
> e(ll) = -118.1067415323327
> e(ll_0) = -140.8673454765096
> e(rank) = 2
>
>
> ----------------------------------------
>> Date: Thu, 14 Apr 2011 15:12:34 +0100
>> Subject: Re: st: list coefficient estimations
>> From: [email protected]
>> To: [email protected]
>>
>> I don't what that means. Perhaps you mean a large number of decimal
>> places. e-class results typically contain far more decimal places than
>> anyone really needs or even wants.
>>
>> On Thu, Apr 14, 2011 at 3:08 PM, Max Fotbollen wrote:
>>
>> > As in SPSS you may get the full decimalpoints for the estimations. That is what I need, full decimalpoints estimations.
>>
>> From: [email protected]
>>
>> >> On Thu, Apr 14, 2011 at 2:40 PM, Max Fotbollen wrote:
>> >> > Hi,
>> >> > I need to retrive and also save OLS estimates. I use return list, don't get any resopnses, not even error response. Use rreturn list, then I get some results in the output, but not the regression coefficients.
>> >> >
>> >>
>> >> What was wrong with the suggestions from Nick Cox?
>> >>
>> >> http://www.stata.com/statalist/archive/2011-04/msg00709.html
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/