Skip to content

Instantly share code, notes, and snippets.

@natematias

natematias/model_results.txt

Last active October 26, 2016 02:46
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save natematias/7af4f44265fd0aa6904be53e0fdf582b to your computer and use it in GitHub Desktop.
Save natematias/7af4f44265fd0aa6904be53e0fdf582b to your computer and use it in GitHub Desktop.
Download ZIP
Statistical modeling of news related comments on a subreddit
Raw
model_results.txt
MSM: ("bbc.com", "reuters.com", "nytimes.com", "washingtonpost.com", "cnn.com",
"telegraph.co.uk", "latimes.com", "huffingtonpost.com", "theguardian.com", "forbes.com",
"examiner.com", "usatoday.com", "wsj.com", "cbsnews.com", "cbc.ca", "time.com",
"sfgate.com", "newsweek.com", "bostonglobe.com", "nydailynews.com", "msnbc.com",
"foxnews.com", "aljazeera.com", "nbcnews.com", "npr.org", "bloomberg.com", "abcnews.com",
"aljazeera.com", "bigstory.ap.com", "cbc.ca", "time.com")
TABLOIDS: ['dailymail.co.uk', 'express.co.uk','mirror.co.uk',
'news.com.au', 'nypost.com', 'thesun.co.uk','dailystar.co.uk','metro.co.uk']
NEGATIVE BINOMIAL MODEL OF THE NUMBER OF COMMENTS WITH LINKS:
=======================================================================
Dependent variable:
-----------------------------------------------------
comment_links
(1) (2) (3)
-----------------------------------------------------------------------
msm 0.463***
(0.059)
tabloid 0.394* 0.521**
(0.160) (0.160)
Constant 0.115*** 0.104*** -0.023
(0.025) (0.025) (0.028)
-----------------------------------------------------------------------
Observations 54,175 54,175 54,175
Log Likelihood -34,104.190 -34,100.770 -34,067.230
theta 0.032*** (0.0005) 0.032*** (0.0005) 0.032*** (0.0005)
Akaike Inf. Crit. 68,210.380 68,205.530 68,140.470
=======================================================================
Note: *p<0.05; **p<0.01; ***p<0.001
NEGATIVE BINOMIAL MODEL OF THE NUMBER OF COMMENTS:
====================================================================
Dependent variable:
--------------------------------------------------
num_comments
(1) (2) (3)
--------------------------------------------------------------------
msm 0.475***
(0.029)
tabloid 0.440*** 0.571***
(0.078) (0.078)
Constant 3.112*** 3.098*** 2.968***
(0.012) (0.012) (0.014)
--------------------------------------------------------------------
Observations 54,175 54,175 54,175
Log Likelihood -135,720.400 -135,702.200 -135,553.700
theta 0.130*** (0.001) 0.130*** (0.001) 0.131*** (0.001)
Akaike Inf. Crit. 271,442.800 271,408.400 271,113.300
====================================================================
Note: *p<0.05; **p<0.01; ***p<0.001
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment