logistic regression - R glm object and prediction using offsets -

-

so i'm using r logistic regression, i'm using offsets.

  mylogit <- glm(y ~  x1 + offset(0.2*x2) + offset(0.4*x3), data = test, family = "binomial")

the output, shows single coefficient, intercept , 1 of predictors, x1.

    coefficients:     (intercept)               x1       0.5250748         0.0157259

my question: how raw prediction each observation model? more specifically, if use predict function, include features , coefficients, though model coefficients listed containing intercept , x1?

  prob = predict(mylogit,test,type=c("response"))

do have use predict function? "mylogit" object contain can compute directly from? (yes looked @ documentation on glm, still confused).

thank patients.

i can report results of experiments glm , offset(). not appear (at least experiment) call predict give results take offset account. rather seems summary.glm needed purpose. started rather mangled modification of 1st example in ?glm ( , more pertinent concerns if did provide data, because play around more newdata argument need "test".)

counts <- c(18,17,15,20,10,20,25,13,12) outcome <- gl(3,1,9) treatment <- gl(3,3) print(d.ad <- data.frame(treatment, outcome, counts)) glm.d93 <- glm(counts ~ outcome + treatment + offset(1:9), family = poisson()) glm.d93d <- glm(counts ~ outcome + treatment , family = poisson())  > predict(glm.d93d, type="response")        1        2        3        4        5        6        7        8        9  21.00000 13.33333 15.66667 21.00000 13.33333 15.66667 21.00000 13.33333 15.66667  > predict(glm.d93, type="response")        1        2        3        4        5        6        7        8        9  21.00000 13.33333 15.66667 21.00000 13.33333 15.66667 21.00000 13.33333 15.66667

as far can tell offset apparent when comparisons of estimated coefficients made null estimate (usually 0) purposes of statistical inference. done summary.glm:

> summary(glm.d93)$coef              estimate std. error    z value      pr(>|z|) (intercept)  2.044522  0.1708987  11.963362  5.527764e-33 outcome2    -1.454255  0.2021708  -7.193203  6.328878e-13 outcome3    -2.292987  0.1927423 -11.896644  1.232021e-32 treatment2  -3.000000  0.2000000 -15.000000  7.341915e-51 treatment3  -6.000000  0.2000000 -30.000000 9.813361e-198 > summary(glm.d93d)$coef                  estimate std. error       z value     pr(>|z|) (intercept)  3.044522e+00  0.1708987  1.781478e+01 5.426767e-71 outcome2    -4.542553e-01  0.2021708 -2.246889e+00 2.464711e-02 outcome3    -2.929871e-01  0.1927423 -1.520097e+00 1.284865e-01 treatment2   1.337909e-15  0.2000000  6.689547e-15 1.000000e+00 treatment3   1.421085e-15  0.2000000  7.105427e-15 1.000000e+00

the offset changing reference levels (with bizarre changes in mangled example) while fitting of $linear.predictors , $fitted data not affected. didn't see comment in glm affects there comment in ?lm : "offsets specified offset not included in predictions predict.lm, whereas specified offset term in formula be." admit got little insight reading ?model.offset.


Comments

Post a Comment

Popular posts from this blog

C# random value from dictionary and tuple -

-
i have following code, , need random item contains string "region_1" dictionary. public static dictionary<int, tuple<string, int, currencytype>> itemarray = new dictionary<int, tuple<string, int, currencytype>>() { { 0xb3e, new tuple<string, int, currencytype>("region_1", 1500, currencytype.fame) } }; public static int generateitemid(string shopid) { var generateditem = itemarray.elementat(new random().next(0, itemarray.count)).key; } how select this? it's not possible efficiently... first create static random @ class level... prevent non-random behaviour if run query on short period of time... (it's seeded discrete clock) static random rnd = new random(); then: var item = itemarray.values .where(t => t.item1 == shopid) .orderby(_ => rnd.next()) .firstordefault()
Read more

.htaccess - htaccess convert request to clean url and add slash at the end of the url -

-
ok , searched more 4 days , more 20 posts here in stach or in websites without kind of success first here htaccess file rewriteengine on rewriterule ^([a-za-z0-9_]+)/?$ index.php?page=$1 [nc] rewriterule ^([a-za-z0-9_]+)\/([a-za-z0-9_]+)\/?$ index.php?page=$1&action=$2 [nc] i tried many codes 1 worked me can call request , works me www.example.com/pagename => www.example.com/index.php?page=pagename here questions , please provide me information "in detail" understand code well 1- how convert request automatically www.example.com/index.php?page=pagename www.example.com/pagename 2- if tried add slash @ end shows me page without images or styling mean www.example.com/pagename => works www.example.com/pagename/ => doesn't work that's :) for first part need 301 rule this: rewritecond %{the_request} /index\.php\?page=([^\s&]+) [nc] rewriterule ^ /%1? [r=301,l] for second part use absolute path in css, js, images files r...
Read more

cgi - How do I interpret URLs without extension as files rather than missing directories in nginx? -

-
so i'm trying install zoneminder , have run under nginx, has few compiled files need run using fast-cgi. these files lack extension. if file has extension, there no issue , nginx interprets file , return 404 if can't find it. if has no extension, assume it's directory , return 404 when can't find sort of index page. here have now: # security camera domain. server { listen 888; server_name mydomain.com; root /srv/http/zm; # enable php support. include php.conf; location / { index index.html index.htm index.php; } # enable cgi support. location ~ ^/cgi-bin/(.+)$ { alias /srv/cgi-bin/zm/; fastcgi_pass unix:/run/fcgiwrap.sock; fastcgi_param script_filename $document_root/$1; include fastcgi.conf; } } the idea is, under cgi-bin directory goes through fastcgi pipe. any idea on howi can fix this? upon closer inspection, realized zoneminder has 2 cgi sc...
Read more