Jonas Moss JonasMoss

## SSC_posting_and_politics_beta.R
## Checks for associations between politics and posting at SSC, using Beta-Bayes!
## Data from: http://slatestarcodex.com/Stuff/ssc2018public.xlsx

library("tidyverse")
SSC = read_excel("ssc2018public.xlsx") # Downloaded from SSC.

Politics = as.factor(SSC$PoliticalAffiliation)
Comments = as.factor(SSC$Comment)

## Not everyone will agree with my classification.

## SSC_posting_and_politics.R
## Makes a plot checking for clear associations between politics and postign in SSC.
## Data from: http://slatestarcodex.com/Stuff/ssc2018public.xlsx

library("tidyverse")
SSC = read_excel("ssc2018public.xlsx") # Downloaded from SSC.

Politics = as.factor(SSC$PoliticalAffiliation)
Comments = as.factor(SSC$Comment)

## Not everyone will agree with my classification.

## trust_simulation.R
N        = 1000
K        = 1
res      = MCMCpack::rdirichlet(N, K*c(0.1, 0.15, 0.7, 0.05))
score    = rowSums(res^2)

trust    = res%*%c(-1,-1,1,0) + rnorm(N, 0, 0.1)
covs     = data.frame(afro = res[1,], latios = res[2,], whites = res[3,], asians = res[4,], score = score)

lm(trust ~ ., data = covs)

## selection_for_significance.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                JonasMoss
                / selection_for_significance.ipynb
            
            
              Last active
              December 4, 2017 16:06
            
              
                Selection for significance.
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## skewness.R
## Needed to calculate moments.
library("moments")
set.seed(313)

## An illustration with both variables skewed.
n = 200
x = rbeta(n, 1, 15)
y = 3 + x + rnorm(n, 0, 0.01)

## Plotting!

## tmvtnorm_correlation_EGD.R
## This script uses the tmvtnorm package, check it out.

if(!("tmvtnorm" %in% rownames(installed.packages()))) {
  install.packages("tmvtnorm")
}

library("tmvtnorm")

## Seed for reproducibility.
set.seed(313)
	## Checks for associations between politics and posting at SSC, using Beta-Bayes!
	## Data from: http://slatestarcodex.com/Stuff/ssc2018public.xlsx

	library("tidyverse")
	SSC = read_excel("ssc2018public.xlsx") # Downloaded from SSC.

	Politics = as.factor(SSC$PoliticalAffiliation)
	Comments = as.factor(SSC$Comment)

	## Not everyone will agree with my classification.
	## Makes a plot checking for clear associations between politics and postign in SSC.
	## Data from: http://slatestarcodex.com/Stuff/ssc2018public.xlsx

	library("tidyverse")
	SSC = read_excel("ssc2018public.xlsx") # Downloaded from SSC.

	Politics = as.factor(SSC$PoliticalAffiliation)
	Comments = as.factor(SSC$Comment)

	## Not everyone will agree with my classification.
	N = 1000
	K = 1
	res = MCMCpack::rdirichlet(N, K*c(0.1, 0.15, 0.7, 0.05))
	score = rowSums(res^2)

	trust = res%*%c(-1,-1,1,0) + rnorm(N, 0, 0.1)
	covs = data.frame(afro = res[1,], latios = res[2,], whites = res[3,], asians = res[4,], score = score)

	lm(trust ~ ., data = covs)
	## Needed to calculate moments.
	library("moments")
	set.seed(313)

	## An illustration with both variables skewed.
	n = 200
	x = rbeta(n, 1, 15)
	y = 3 + x + rnorm(n, 0, 0.01)

	## Plotting!
	## This script uses the tmvtnorm package, check it out.

	if(!("tmvtnorm" %in% rownames(installed.packages()))) {
	install.packages("tmvtnorm")
	}

	library("tmvtnorm")

	## Seed for reproducibility.
	set.seed(313)