kosugitti

data{
  int<lower=0> N;
  vector[2] X[N];
}
parameters{
  vector[2] mu;
  vector<lower=0>[2] sigma;
  real<lower=-1,upper=1> rho;
}
transformed parameters{

kosugitti

data{
 int<lower=0> N;
 vector[3] X[N];
}
parameters{
  vector[3] mu;
  real<lower=0> sig[3];
  real<lower=-1,upper=1> rho;
}
transformed parameters{

kosugitti

data{
 int<lower=0> N;
 real X1[N];
 real X2[N];
 real X3[N];
}
parameters{
  real mu1;
  real mu2;
  real mu3;

kosugitti

data{
 int<lower=0> N;
 vector[2] X1[N];
 vector[2] X2[N];
}
parameters{
  real mu;
  real<lower=0> sig[2];
  real<lower=-1,upper=1> rho;
  real effectA; //between

kosugitti

library(rvest)
p.data <- read_html("http://baseball-data.com/ranking-salary/all/p.html") #ピッチャーの年俸など
h.data <- read_html("http://baseball-data.com/ranking-salary/all/h.html") #野手のそれ
avg5 <- read_html("http://baseball-data.com/stats/hitter2-all/avg-5.html") #規定打数1/3以上の打者基本成績
era5 <- read_html("http://baseball-data.com/stats/pitcher2-all/era-5.html") #規定投球回1/3以上の投手基本成績
p.df <- as.data.frame(html_table(p.data,head=T))
head(p.df)

library(stringr) # 文字列操作パッケージ
p.df$順位 <- NULL

kosugitti

library(rstan)
library(ggplot2)
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())
N <- 10000
mu <- 10
sigma <- 20
X <- rnorm(N,mu,sigma)

modelcode <- '

kosugitti

# 囚人のジレンマ関数をつくる
PDG.rule <- function(A,B){
  if(A ==1 && B == 1){ payA <-  8; payB <-  8 }
  if(A ==0 && B == 1){ payA <- 12; payB <-  0 }
  if(A ==1 && B == 0){ payA <-  0; payB <- 12 }
  if(A ==0 && B == 0){ payA <-  4; payB <-  4 }
  return(list(A=payA,B=payB))
}

PDG.rule(0,1)

kosugitti

# 初期設定
sizeX <- 20
sizeY <- 20
timeMax <- 60
# dat <- matrix(c(0,0,0,0,0,0,
#                 0,0,1,0,0,0,
#                 0,0,0,1,0,0,
#                 0,1,1,1,0,0,
#                 0,0,0,0,0,0,
#                 0,0,0,0,0,0),nrow=6,ncol=6,byrow=T)

kosugitti

cellsize <- 160
trace <- 50

cell <- matrix(ncol=cellsize,nrow=trace)
# 初期化
cell[1,] <- rbinom(cellsize,1,0.5)

### into 2bit
dec2bin <- function(num,digit=8){
  if(num <= 0 && digit <= 0){

kosugitti

require(mvtnorm)

imKmeans <- function(dat,cl){
  # step 1
  dat.km <- kmeans(dat,cl,algorithm = "MacQueen")
  # step 2
  conv <- 3
  clust <- dat.km$cluster

  while(conv>0){