S-Katagiri Gedevan-Aleksizde

## Rstan_VARMApq.R
library(tidyr)    # ver. 0.3.1.
library(dplyr)    # ver. 0.4.3
library(rstan)    # ver. 2.9.0
library(forecast) # ver. 6.2

# 並列化させる設定
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())

N <- 2              # num series

## varma_pop.R
library(tidyr)    # ver. 0.3.1.
library(dplyr)    # ver. 0.4.3
library(rstan)    # ver. 2.9.0
library(forecast) # ver. 6.2
library(zoo)      # ver. 1.7-11
library(gdata)    # ver. 2.13.3
library(loo)      # ver. 0.1.6


## Rstan_VARMA11_2.R
library(tidyr)    # ver. 0.3.1.
library(dplyr)    # ver. 0.4.3
library(rstan)    # ver. 2.9.0
library(forecast) # ver. 6.2
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())

# VARMA(1,1) N=2
N <- 2              # num series
Time <- 400         # span

## Gaussian_VARMA.stan
/* VARMA (p,q) */

data {
	int<lower=1> T ; // num observations
	int<lower=1> N ; // num series
	int<lower=0> p ; // AR(p)
	int<lower=0> q ; // MA(q)
	vector[N] y[T] ; // observed outputs
	int<lower=0> T_forecast ; // forecasting span
}

## Rstan_VARMA11.R
library(tidyr)    # ver. 0.3.1.
library(dplyr)    # ver. 0.4.3
library(rstan)    # ver. 2.9.0
library(forecast) # ver. 6.2

# 並列化させる設定
rstan_options(auto_write = TRUE)
options(mc.cores = parallel::detectCores())

setwd("~/Documents/blog/20160212_StanVARMA/") # 任意のフォルダに書き換える

## VARMApq.stan
/* VARMA (p,q) */

data {
	int<lower=1> T ; // num observations
	int<lower=1> N ; // num series
	int<lower=0> p ; // AR(p)
	int<lower=0> q ; // MA(q)
	vector[N] y[T] ; // observed outputs
	int<lower=0> T_forecast ; // forecasting span
}

## VARMA11.stan
/* VARMA (1,1) */

data {
	int<lower=1> T ; // num observations
	int<lower=1> N ; // num series
	vector[N] y[T] ; // observed outputs
	int<lower=0> T_forecast ; // forecasting span
}

parameters {

## RFM_Bayes.stan
/* -----------------------------------------------------------------------------
Estimate latent variable hierarchical Bayes RFM model

stan ver. 2.9

CREATED BY: 09/APR/2015, ill-identified
------------------------------------------------------------------------------- */

data {
  int<lower=1>              N ; // num of customers

## RFM_stan_exec.R
library(dplyr)
library(tidyr)
library(ggplot2)
library(rstan)
library(loo)

# working directory
work.dir <- "HOGEHOGE/20160406_RFM"

df <- read.csv(paste(work.dir, "rfm.csv",sep="/"), stringsAsFactors = F)

## rfm_handle.py
# !/usr/bin/env python3
# -*- coding: utf-8 -*-
# coding: utf-8

# In[1]:

get_ipython().magic('matplotlib inline'
import pystan        # ver. 2.9.0.0
import numpy  as np  # ver.1.11.0
import pandas as pd  # ver. 0.18.0 # xlrd, mumxpr module needed