Discussion:
Odds of tRUMP & Dems Rising & Falling _Together_ In the Polls
Add Reply
Bret Cahill
2018-08-31 01:03:22 UTC
Reply
Permalink
Raw Message
Every Monday Gallup publishes tRUMP's approval for the previous week.

https://news.gallup.com/poll/203207/trump-job-approval-weekly.aspx

A meta poll on the 2018 generic Democrat is updated more often but it also has a lag as well.

https://www.realclearpolitics.com/epolls/other/2018_generic_congressional_vote-6185.html

Ignore the lag and work week by week back in time to get a correlation coefficient between the +/- changes in tRUMP's approval and the +/- changes in the generic Dem.

You would think that after you go back a few weeks a negative coefficient would start to appear. News good for tRUMP would be bad for Dems and vice versa.

At a minimum the correlation would tend to zero.

If the coefficient starts to tend to +1 sumthin ain't addin' up and we need to tip off Columbia Journalism Review for a formal study:

https://groups.google.com/forum/#!topic/alt.philosophy/yFSMKiAYnLg

CJR can deliver such a blow the jerryspringer media . . .


Bret Cahill


Odds of tRUMP & Dems Rising & Falling _Together_ In the Polls
Kym Horsell
2018-08-31 01:50:16 UTC
Reply
Permalink
Raw Message
Post by Bret Cahill
Every Monday Gallup publishes tRUMP's approval for the previous week.
https://news.gallup.com/poll/203207/trump-job-approval-weekly.aspx
A meta poll on the 2018 generic Democrat is updated more often but it also has a lag as well.
https://www.realclearpolitics.com/epolls/other/2018_generic_congressional_vote-6185.html
Ignore the lag and work week by week back in time to get a correlation coefficient between the +/- changes in tRUMP's approval and the +/- changes in the generic Dem.
You would think that after you go back a few weeks a negative coefficient would start to appear. News good for tRUMP would be bad for Dems and vice versa.
At a minimum the correlation would tend to zero.
https://groups.google.com/forum/#!topic/alt.philosophy/yFSMKiAYnLg
CJR can deliver such a blow the jerryspringer media . . .
Bret Cahill
Odds of tRUMP & Dems Rising & Falling _Together_ In the Polls
You can allays count on the ficklness and perfidity of humin naycha. :)


TS regression of NYSE:XOM price on Friday and Trump approval ending that weekend
(from above link):

y = 0.278825*x + 18.6112
sdx 4.97532
rss 9.67555
se 1.09975
r 0.40731
limits for beta at 90.0% CI
tc = 1.85955 at 8 d.f.
beta in 0.278825 +- 0.411035 = [-0.132209, 0.68986]
T-tests on beta:
H0 beta == 0.000000 against H1 beta != 0.000000
calculated t = 1.26142 at 8 d.f.
|t| <= tc (1.85955 2-sided); accept H0
H0 beta == 0.000000 against H1 beta > 0.000000
t <= tc (1.39682 right tail); accept H0
Probabilities:
P(beta!=0.000000) = 0.757306
P(beta>0.000000) = 0.878653
limits for alpha at 90.0% CI
tc = 1.85955 at 8 d.f.
alpha in 18.6112 +- 33.3058 = [-14.6946, 51.9171]
r2 = 0.165901
calculated Spearman corr = 0.575758
Testing:
H0: vars are independent
|r| > rc (0.564000 2-sided) at 5%; reject H0
tag x y yp
2018/8/19 78.26 42 40.4321*
2018/8/12 79.08 39 40.6608*
2018/8/26 79.62 41 40.8113
2018/8/05 80.2 41 40.973
2018/6/24 81.38 41 41.3021
2018/7/22 81.4 42 41.3076
2018/7/29 81.92 40 41.4526*
2018/7/08 82.24 41 41.5418
2018/7/01 82.73 42 41.6785
2018/7/15 83.31 43 41.8402*


IOW trump approval (at least) seems significantly related to strength
of random stock on the NYSE aka general optimism metric.
Bret Cahill
2018-09-04 04:23:17 UTC
Reply
Permalink
Raw Message
Post by Kym Horsell
Post by Bret Cahill
Every Monday Gallup publishes tRUMP's approval for the previous week.
https://news.gallup.com/poll/203207/trump-job-approval-weekly.aspx
A meta poll on the 2018 generic Democrat is updated more often but it also has a lag as well.
https://www.realclearpolitics.com/epolls/other/2018_generic_congressional_vote-6185.html
Ignore the lag and work week by week back in time to get a correlation coefficient between the +/- changes in tRUMP's approval and the +/- changes in the generic Dem.
You would think that after you go back a few weeks a negative coefficient would start to appear. News good for tRUMP would be bad for Dems and vice versa.
At a minimum the correlation would tend to zero.
https://groups.google.com/forum/#!topic/alt.philosophy/yFSMKiAYnLg
CJR can deliver such a blow the jerryspringer media . . .
Bret Cahill
Odds of tRUMP & Dems Rising & Falling _Together_ In the Polls
You can allays count on the ficklness and perfidity of humin naycha. :)
There's nothing fickle about a high correlation.

As perdicted, the Dems soared to 8.4% and tRUMP's disapproval dropped a fraction of a point. His approval held steady which is purty good considering all the other bad news for tRUMP.

This keeps up and CJR will jerk a knot in legacy media fanny that will not quit.


Bret Cahill
Bret Cahill
2018-09-11 00:04:26 UTC
Reply
Permalink
Raw Message
Post by Bret Cahill
Post by Kym Horsell
Post by Bret Cahill
Every Monday Gallup publishes tRUMP's approval for the previous week.
https://news.gallup.com/poll/203207/trump-job-approval-weekly.aspx
A meta poll on the 2018 generic Democrat is updated more often but it also has a lag as well.
https://www.realclearpolitics.com/epolls/other/2018_generic_congressional_vote-6185.html
Ignore the lag and work week by week back in time to get a correlation coefficient between the +/- changes in tRUMP's approval and the +/- changes in the generic Dem.
You would think that after you go back a few weeks a negative coefficient would start to appear. News good for tRUMP would be bad for Dems and vice versa.
At a minimum the correlation would tend to zero.
https://groups.google.com/forum/#!topic/alt.philosophy/yFSMKiAYnLg
CJR can deliver such a blow the jerryspringer media . . .
Bret Cahill
Odds of tRUMP & Dems Rising & Falling _Together_ In the Polls
You can allays count on the ficklness and perfidity of humin naycha. :)
There's nothing fickle about a high correlation.
As perdicted, the Dems soared to 8.4% and tRUMP's disapproval dropped a fraction of a point. His approval held steady which is purty good considering all the other bad news for tRUMP.
This keeps up and CJR will jerk a knot in legacy media fanny that will not quit.
tRUMP drops from 41% to 40% as the generic Dem drops from 8.4 to 7.8

https://news.gallup.com/poll/203207/trump-job-approval-weekly.aspx

https://www.realclearpolitics.com/epolls/other/2018_generic_congressional_vote-6185.html
Loading...