css table

.td1_th {border-width:1px;border-color:#e1e1e1;border-style:solid;background-color:#ffffff;}

javascript

var div = document.getElementById("container");
var rect = div.getBoundingClientRect();

document.getElementById("banner").innerHTML = ("<div style='position:absolute;top: "+rect.left+"px'> <img src='http://hmwatch.woobi.co.kr/sci/images/icm.png' />");

x axis date

d

ar1 of kospi

with prediction

library(quantmod)
library(fGarch)

URL <- "http://ichart.finance.yahoo.com/table.csv?s=^KS11"
dat <- read.csv(URL)
dat$Date <- as.Date(dat$Date, "%Y-%m-%d")

odat<-dat[c(1:100),]

attach(odat)

x<-odat[order(Date),]

#kospiGarch <- garchFit(~arma(1, 2) + garch(1, 1), data=as.ts(x$Close))

fit <- arima(diff(log(x$Close)), order=c(1,0,0))

#fore <- predict(fit, n.ahead=10, doplot=F)

err <- rnorm(20, 0, .002)

sim <- diff(log(abs(x$Close)))[99]

for (t in 2:20) {
    sim[t] <- fit$coef[2]+ fit$coef[1]*sim[t-1] + err[t]
}

dev.new()

par(mfrow=c(3,1))
par(mar=c(3,3,1,2))

plot(x$Close, type='l', xlim=c(0, 120), xaxt='n',
    panel.first=grid(col = "gray", lty = "dotted"), pch=20)
t <- c(101:120)
points(x$Close, pch = 20, col = "red")

ye <- x$Close[100]
for (j in 2:20) {
ye[j] <- exp(sim[j-1])*ye[j-1]
}

ye
points(t, ye, col='red', pch=20, type='b')


##########################################################

library(forecast)

fit <- auto.arima(x$Close)
plot(forecast(fit, h=20))

arch(1) model

require(tseries)

URL <- "http://ichart.finance.yahoo.com/table.csv?s=^KS11"
dat <- read.csv(URL)
dat$Date <- as.Date(dat$Date, "%Y-%m-%d")

odat<-dat[c(1:100),]

N<-99

attach(odat)

x<-odat[order(Date),]


price <- x[[2]]
ret <- 0

for (i in 1:length(price)) {
ret[i] <- (price[i]-price[1])/price[1]
}

par(mfrow=c(3,1))
par(mar=c(3,3,1,2))
x=c(1:length(ret))
plot(price, type='l', xaxt='n')

y<-diff(price)
plot(x = y, ylab = expression(y[t]), xlab = "t", type = "l", col = "red", 
     panel.first=grid(col = "gray", lty = "dotted"))
  points(x = y, pch = 20, col = "blue")
 

x.arch <- garch(ret-mean(ret), order = c(0,1), trace=TRUE)  # Fit ARCH(1)

#summary(x.arch)                     # Diagnostic tests
vol <- x.arch$fitted.values[,1]

plot(x, ret, type='l', col='lightgray', ylim=c(-0.01,0.08), xaxt='n')
points(x=x, y=vol, col='red', type='l')

abline(h=0, col="lightgray")

파이를 무한자리수까지 계산하는 파이썬 코드

교양수학책을 읽다보면 파이나 e 같은 무한소수의 자리수를 10조자리까지 계산했다느니 뭐니 하는 얘기를 자주 접하게 된다. 그래서 지금 펜티엄코어2로 돌리면 시간이 어느 정도 걸리나 한번 테스트해보려고 했다.

그런데 막상 c로 알고리즘을 만드려니까 자리수 계산하려면 배열에 동적메모리 할당해서 넣어야 하는데 막상 구현하려니 머리가 멍해짐. 웹페이지 검색해보니 코드와 함께 우리말로 설명이 된 웹페이지는 찾을 수가 없었다.

구글에서 infinite precision pi로 검색하다가 c#으로 된 페이지를 겨우 찾았는데 읽다가 졸려서 아침에 보려고 나뒀는데 실수로 검색창을 닫아버려서 다시 검색하다가 발견한 페이지.


http://rosettacode.org/wiki/Pi

c로 하려니까 gmp.h 헤더파일 인클루드하라고 에러 나와서 귀찾아서 파이선으로 돌림.

파이선 코드
 ------------
def calcPi():
    q, r, t, k, n, l = 1, 0, 1, 1, 3, 3
    while True:
        if 4*q+r-t < n*t:
            yield n
            nr = 10*(r-n*t)
            n  = ((10*(3*q+r))//t)-10*n
            q  *= 10
            r  = nr
        else:
            nr = (2*q+r)*l
            nn = (q*(7*k)+2+(r*l))//(t*l)
            q  *= k
            t  *= l
            l  += 2
            k += 1
            n  = nn
            r  = nr

import sys
pi_digits = calcPi()
i = 0
for d in pi_digits:
    sys.stdout.write(str(d))
    i += 1
    if i == 40: print(""); i = 0

neural network kospi

#install.packages('neuralnet')
library("neuralnet")
#Going to create a neural network to perform sqare rooting
#Type ?neuralnet for more information on the neuralnet library
#Generate 50 random numbers uniformly distributed between 0 and 100
#And store them as a dataframe

library(quantmod)
library(fGarch)

URL <- "http://ichart.finance.yahoo.com/table.csv?s=^KS11"
dat <- read.csv(URL)
dat$Date <- as.Date(dat$Date, "%Y-%m-%d")

odat<-dat[c(1:100),]

attach(odat)

x<-odat[order(Date),]

t <- c(1:100)
traininginput <-  cbind(t, Lag(x$Close, 1))
trainingoutput <- x$Close


#Column bind the data into one variable
trainingdata <- data.frame(traininginput,trainingoutput)
colnames(trainingdata) <- c("Input","Output")


#Train the neural network
#Going to have 10 hidden layers
#Threshold is a numeric value specifying the threshold for the partial
#derivatives of the error function as stopping criteria.
net.sqrt <- neuralnet(Output~Input,trainingdata, hidden=5, threshold=0.01)

#Plot the neural network
plot(net.sqrt)

#Test the neural network on some training data
testdata <- c(101:110) #Generate some squared numbers
net.results <- compute(net.sqrt, testdata) #Run them through the neural network

#Lets display a better version of the results
cleanoutput <- cbind(testdata,sqrt(testdata),
                         as.data.frame(net.results$net.result))
colnames(cleanoutput) <- c("Input","Expected Output","Neural Net Output")


cleanoutput