y[n] y[n 1], y[n 2] v[n] R k c m y(t) Figure 1: series.r series <- function(n,delta,k,c,m,y0,y1) { a1 <- (2* m/delta^2 - k) / (m/delta^ *c/delta
|
|
|
- ひろみ おおふさ
- 7 years ago
- Views:
Transcription
1 ARMA 1 (AIC) ARMA 2 ARMA dy(t) = f(t, y), y(0) = y 0 dt f(t, y) y 0 f(t, y) LCR (Fig. 1) k c m t y(t) F = ma F = θ(t) ky(t) c dy(t), ma = m d2 y(t) dt dt 2 m d2 y(t) dt 2 + c dy(t) dy(0) + ky(t) = v(t), y(0) = y 0, = y 1 dt dt v(t) y 0 Y 1 m v(t) v(t) y(t) y(t) y[n] y[n] 2y[n 1] + y[n 2] y[n] y[n 2] m 2 + c + ky[n 1] = v[n] 2 Delta y[n] = a 1 y[n 1] + a 2 y[n 2] + b 0 v[n] (n = 0, 1, 2,, N 1) ( ) /( 2m m a 1 = 2 k 2 + c ) 2 ( a 2 = m 2 + c ) /( m c ) 2 /( m b 0 = c ) 2
![= y 0, = y 1 dt dt v(t) y 0 Y 1 m v(t) v(t) y(t) y(t) y[n] y[n] 2y[n 1] + y[n 2] y[n] y[n 2] m 2 + c + ky[n 1] = v[n] 2](/docs-images/46/20539321/images/page_1.jpg "Delta y[n] = a 1 y[n 1] + a 2 y[n 2] + b 0 v[n] (n = 0, 1, 2,, N 1) ( ) /( 2m m a 1 = 2 k 2 + c ) 2 ( a 2 = m 2 + c ) /(")
2 y[n] y[n 1], y[n 2] v[n] R k c m y(t) Figure 1: series.r series <- function(n,delta,k,c,m,y0,y1) { a1 <- (2* m/delta^2 - k) / (m/delta^ *c/delta); a2 <- (- m/delta^ *c/delta) / (m/delta^ *c/delta); b0 <- 1 / (m/delta^ *c/delta); y <- numeric(n+2); v <- numeric(n); # for simulation 1 #v <- rnorm(n); # for simulation 2 y[1] <- y0; y[2] <- y[1] + y1*delta; list = 1:n; for(i in list) { y[i+2] <- a1*y[i+1] + a2*y[i] + v[i] return(y); Fig. 2 N = 256, = 1, k = 0.2, c = 0.06, m = 1, y0 = 1, y1 = 0 v[n] = 0 (n = 0, 1,, N 1)
3 (1) > source("series.r") > y <- series(256,1,0.2,0.06,1,1,0) > plot(y, type="l") > dev.copy2eps(file="simulation1.eps", width=10, height=6) y Figure 2: (1) (v[n] = 0) y 0 = y 1 = 0 Fig. 3 (1) > source("series.r") > y <- series(256,1,0.2,0.06,1,0,0) > plot(y, type="l") > dev.copy2eps(file="simulation2.eps", width=10, height=6 )
4 y Figure 3: (2) 3 ARMA y[n] v[n] y[n] = a i y[n i] + v[n] b i v[n i] ARMA (AutoRegressive Moving Average model) m, {a i m (AR coefficients) l, {b i l (MA coefficients) v[n] 0 σ 2 y[n i] v[n] E{v[n] = 0, E{v[n]v[m] = σ 2 δ mn, E{v[n]y[m] = 0 (n > m) δ mn { 1 (n = m) δ mn = 0 Otherwise ARMA l = 0 y[n] = a i y[n i] + v[n] m (AutoRegressive Model) AR m = 0 (Moving Average Model) MA y[n] = v[n] b i v[n i]
![E{v[n] = 0, E{v[n]v[m] = σ 2 δ mn, E{v[n]y[m] = 0 (n > m) δ mn { 1 (n = m) δ mn = 0 Otherwise ARMA l =](/docs-images/46/20539321/images/page_4.jpg "0 y[n] = a i y[n i] + v[n] m (AutoRegressive Model) AR m = 0 (Moving Average Model) MA y[n] = v[n] b i")
5 (a) 2 AR y[n] = 0.9 3y[n 1] 0.81y[n 2] + v[n] (b) 2 MA y[n] = v[n] 0.9 2v[n 1] v[n 2] (c) (2, 2) ARMA y[n] = 0.9 3y[n 1] 0.81y[n 2] + v[n] 0.9 2v[n 1] v[n 2] R > n <- 256 > m <- 2 > a <- c(0.9*sqrt(3), -0.81) > l <- 2 > b <- c(0.9*sqrt(2), -0.81) >sigma2 <- 1 # AR(2) model > source("armodel.r") > y <- armodel(n,m,a,sigma2) > plot(y, type="l") > dev.copy2eps(file="armodel.eps", width=10, height=6) # MA(2) model > source("mamodel.r") > y <- mamodel(n,l,b,sigma2) > plot(y, type="l") > dev.copy2eps(file="mamodel.eps", width=10, height=6) # ARMA(2,2) model > source("armamodel.r") > y <- armamodel(n,m,l,a,b,sigma2) > plot(y, type="l") > dev.copy2eps(file="armamodel.eps", width=10, height=6)
6 ARmodel.R # AR model armodel <- function(n,m,a,sigma2) { y <- numeric(n+m); v <- rnorm(n, mean=0, sd=sqrt(sigma2)); listi = 1:n; listj = 1:m; for(i in listi) { y[i+m] <- v[i]; for(j in listj) { y[i+m] <- y[i+m] + a[j] * y[i+m-j]; y <- c(y)[m+1:n]; return(y); y Figure 4: (AR model)
7 MAmodel.R # MA model mamodel <- function(n,l,b,sigma2) { y <- numeric(n); v <- c(numeric(l),rnorm(n, mean=0, sd=sqrt(sigma2))); listi = 1:n; listj = 1:l; for(i in listi) { y[i] <- v[i+l]; for(j in listj) { y[i] <- y[i] - b[j] * v[i+l-j]; return(y); y Figure 5: (MA model)
8 ARMAmodel.R # ARMA model armamodel <- function(n,m,l,a,b,sigma2) { y <- numeric(n+m); v <- c(numeric(l),rnorm(n, mean=0, sd = sqrt(sigma2))); listi = 1:n; listj = 1:l; listk = 1:m; for(i in listi) { y[i+m] <- v[i+l]; for(j in listj) { y[i+m] <- y[i+m] - b[j] * v[i+l-j]; for (k in listk) { y[i+m] <- y[i+m] + a[k] * y[i+m-k]; y <- c(y)[m+1:n]; return(y); y Figure 6: (ARMA model)
9 4 ARMA ARMA z 1 D ARMA ( ) ( ) 1 a i z i y[n] = 1 b i z i v[n] AR MA ( ) a(z 1 ) := 1 a i z i ARMA a(z 1 )y[n] = b(z 1 )v[n] ( ) b(z 1 ) := 1 b i z i z 1 g(z 1 ) = { a(z 1 ) 1 b(z 1 ) = g i z i i=0 y[n] = g(z 1 )v[n] = g i v[n i] i=0 ARMA MA g i ARMA i g 0 = 1, g i = a j g i j b i (i > 1) j=1 AR MA a j = 0 (j > m) b i = 0 (i > l) ARMA 5 ARMA ARMA y[n k] E(y[n]y[n k]) = a i E(y[n i]y[n k]) + E(v[n]y[n k]) b i E(v[n i]y[n k]) ARMA MA { 0 (n > m) E(v[n]y[m]) = g i E(v[n]v[m i]) = σ 2 g m n (n m) i=0
![i g 0 = 1, g i = a j g i j b i (i > 1) j=1 AR MA a j = 0 (j > m) b i = 0 (i > l) ARMA 5 ARMA ARMA y[n k] E(y[n]y[n k]) = a i](/docs-images/46/20539321/images/page_9.jpg "E(y[n i]y[n k]) + E(v[n]y[n k]) b i E(v[n i]y[n k]) ARMA MA { 0 (n > m) E(v[n]y[m]) = g i E(v[n]v[m i]) = σ 2 g m n (n m) i=0")
10 ARMA C 0 = C k = ) a i C i + σ (1 2 b i g i a i C k i σ 2 b i g i k (k = 1, 2, ) ARMA AR MA {m, a 1, a 2,, a m, {l, b 1, b 2,, b l σ 2 {g i,2, {C k k=0,1,2, AR C 0 = C k = a i C i + σ 2 a i C k i (k = 1, 2, ) Yule- Walker AR 6 ARMA ARMA C k (DTFT) p(f) = DTFT{C k = k= C k exp( 2πjkf) =< C k, exp(2πjkf) > ( 1 2 f 1 ) 2 C k (power spectral density function) (Spectrum) p(f) = = = = k= C k exp( 2πjkf) E(y[n]y[n k]) exp( 2πjkf) k= {( ) ( ) E g r v[n r] g s v[n k s] exp( 2πjkf) k= r=0 s=0 g r g s E(v[n r]v[n k s]) exp( 2πjkf) k= r=0 s=0 g r = 0 (r < 0) p(f) = σ 2 g r g r k exp( 2πjkf) k= r=0
![exp(2πjkf) > ( 1 2 f 1 ) 2 C k (power spectral density function) (Spectrum) p(f) = = = = k= C k exp( 2πjkf) E(y[n]y[n k]) exp( 2πjkf) k= {( ) ( ) E](/docs-images/46/20539321/images/page_10.jpg "g r v[n r] g s v[n k s] exp( 2πjkf) k= r=0 s=0 g r g s E(v[n r]v[n k s]) exp( 2πjkf) k= r=0 s=0 g r = 0 (r < 0) p(f) = σ 2 g r g r k exp( 2πjkf) k=")
11 = σ 2 r g r exp( 2πjrf)g r k exp{ 2πj(k r)f r=0 k= = σ 2 g r exp( 2πjrf)g s exp( 2πjsf) = r=0 s=0 σ 2 2 g r exp( 2πjrf) r=0 z = exp(2πjf) g(z 1 ) { 1 { exp( 2πjrf) = 1 a i exp( 2πjif) 1 b i exp( 2πjif) r=0 ARMA 1 l p(f) = σ 2 b i exp( 2πjif) 2 1 m a i exp( 2πjif) 2 ARMA AR {a i m MA {b i l σ 2 AR(1),AR(2),MA(1) ARMA(1,1) AR MA ARMA R plot(y, type= l ) y acf(y) y spec.pgram(y, type= h ) y AR(2) MA(2) ARMA(2,2)
12 y 5 5 Series y ACF Lag Series: y Raw Periodogram spectrum 1e frequency bandwidth = Figure 7: AR model y Series y ACF Lag Series: y Raw Periodogram spectrum 1e 03 1e 01 1e frequency bandwidth = Figure 8: MA model
13 y Series y ACF Lag spectrum 1e 03 1e 01 1e+01 Series: y Raw Periodogram frequency bandwidth = Figure 9: ARMA model
arma dvi
ARMA 007/05/0 Rev.0 007/05/ Rev.0 007/07/7 3. : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 3. : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 3.3 : : : :
…K…E…X„^…x…C…W…A…fi…l…b…g…‘†[…N‡Ì“‚¢−w‘K‡Ì‹ê™v’«‡É‡Â‡¢‡Ä
2009 8 26 1 2 3 ARMA 4 BN 5 BN 6 (Ω, F, µ) Ω: F Ω σ 1 Ω, ϕ F 2 A, B F = A B, A B, A\B F F µ F 1 µ(ϕ) = 0 2 A F = µ(a) 0 3 A, B F, A B = ϕ = µ(a B) = µ(a) + µ(b) µ(ω) = 1 X : µ X : X x 1,, x n X (Ω) x 1,,
-- Blackman-Tukey FFT MEM Blackman-Tukey MEM MEM MEM MEM Singular Spectrum Analysis Multi-Taper Method (Matlab pmtm) 3... y(t) (Fourier transform) t=
--... 3..... 3...... 3...... 3..3....3 3..4....4 3..5....5 3.....6 3......6 3......7 3..3....0 3..4. Matlab... 3.3....3 3.3.....3 3.3.....4 3.3.3....4 3.3.4....5 3.3.5....5 3.4. MEM...8 3.4.. MEM...8 3.4..
Recent Developments and Perspectives of Statistical Time Series Analysis /ta) : t"i,,t Q) w (^ - p) dp *+*ffi t 1 ] Abraham, B. and Ledolter, J. (1986). Forecast functions implied by autoregressive
<4D F736F F D B B83578B6594BB2D834A836F815B82D082C88C60202E646F63>
通信方式第 2 版 サンプルページ この本の定価 判型などは, 以下の URL からご覧いただけます. http://www.morikita.co.jp/books/mid/072662 このサンプルページの内容は, 第 2 版発行当時のものです. i 2 2 2 2012 5 ii,.,,,,,,.,.,,,,,.,,.,,..,,,,.,,.,.,,.,,.. 1990 5 iii 1 1
橡表紙参照.PDF
CIRJE-J-58 X-12-ARIMA 2000 : 2001 6 How to use X-12-ARIMA2000 when you must: A Case Study of Hojinkigyo-Tokei Naoto Kunitomo Faculty of Economics, The University of Tokyo Abstract: We illustrate how to
Stata 11 Stata ts (ARMA) ARCH/GARCH whitepaper mwp 3 mwp-083 arch ARCH 11 mwp-051 arch postestimation 27 mwp-056 arima ARMA 35 mwp-003 arima postestim
TS001 Stata 11 Stata ts (ARMA) ARCH/GARCH whitepaper mwp 3 mwp-083 arch ARCH 11 mwp-051 arch postestimation 27 mwp-056 arima ARMA 35 mwp-003 arima postestimation 49 mwp-055 corrgram/ac/pac 56 mwp-009 dfgls
aisatu.pdf
1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
... 3... 3... 3... 3... 4... 7... 10... 10... 11... 12... 12... 13... 14... 15... 18... 19... 20... 22... 22... 23 2
1 ... 3... 3... 3... 3... 4... 7... 10... 10... 11... 12... 12... 13... 14... 15... 18... 19... 20... 22... 22... 23 2 3 4 5 6 7 8 9 Excel2007 10 Excel2007 11 12 13 - 14 15 16 17 18 19 20 21 22 Excel2007
[1] 1.1 x(t) t x(t + n ) = x(t) (n = 1,, 3, ) { x(t) : : 1 [ /, /] 1 x(t) = a + a 1 cos πt + a cos 4πt + + a n cos nπt + + b 1 sin πt + b sin 4πt = a
![[1] 1.1 x(t) t x(t + n ) = x(t) (n = 1,, 3, ) { x(t) : : 1 [ /, /] 1 x(t) = a + a 1 cos πt + a cos 4πt + + a n cos nπt + + b 1 sin πt + b sin 4πt = a](/thumbs/85/91634532.jpg "[1] 1.1 x(t) t x(t + n ) = x(t) (n = 1,, 3, ) { x(t) : : 1 [ /, /] 1 x(t) = a + a 1 cos πt + a cos 4πt + + a n cos nπt + + b 1 sin πt + b sin 4πt = a")
13/7/1 II ( / A: ) (1) 1 [] (, ) ( ) ( ) ( ) etc. etc. 1. 1 [1] 1.1 x(t) t x(t + n ) = x(t) (n = 1,, 3, ) { x(t) : : 1 [ /, /] 1 x(t) = a + a 1 cos πt + a cos 4πt + + a n cos nπt + + b 1 sin πt + b sin
http://banso.cocolog-nifty.com/ 100 100 250 5 1 1 http://www.banso.com/ 2009 5 2 10 http://www.banso.com/ 2009 5 2 http://www.banso.com/ 2009 5 2 http://www.banso.com/ < /> < /> / http://www.banso.com/
パソコン接続マニュアル P-01F 日本語
P-01F 1 2 3 4 5 1 2 +m1111 1 2 3 4 5 6 6 1 1 111 2 1 3 1 1 1 2 1 7 3 8 1 2 1 111 3 4 5 9 1 m111 m1111 c 2 3 4 5 10 1 1 111 2 1 3 1 1 1 2 1 3 1 m111 m1111 2 3 1 11 12 1 2 3 1 2 3 13 1 2 3 4 5 14 6 7 8 9
GT-3 P GT-3-6F GT GT GT GT GT GT GT
150-1.5GT-3 P 30-1.5GT-3-6F 40.5-1.5GT 40.50 27 45-1.5GT 45.00 30 48-1.5GT 48.00 32 51-1.5GT 51.00 34 61.5-1.5GT 61.50 41 69-1.5GT 69.00 46 70.5-1.5GT 70.50 47 72-1.5GT 72.00 48 73.5-1.5GT 73.50 49 76.5-1.5GT
3 3 i
00D8102021I 2004 3 3 3 i 1 ------------------------------------------------------------------------------------------------1 2 ---------------------------------------------------------------------------------------2
n=360 28.6% 34.4% 36.9% n=360 2.5% 17.8% 19.2% n=64 0.8% 0.3% n=69 1.7% 3.6% 0.6% 1.4% 1.9% < > n=218 1.4% 5.6% 3.1% 60.6% 0.6% 6.9% 10.8% 6.4% 10.3% 33.1% 1.4% 3.6% 1.1% 0.0% 3.1% n=360 0% 50%
今回用いる例データ lh( 小文字のエル ) ある女性の血液中の黄体ホルモンを 10 分間隔で測定した時系列データ UKgas 1960 年 ~1986 年のイギリスのガス消費量を四半期ごとに観測した時系列データ ldeaths 1974 年 ~1979 年のイギリスで喘息 気管支炎 肺気腫による死
12 章 - 時系列分析 1296603c 埴岡瞬 今回用いる例データ lh( 小文字のエル ) ある女性の血液中の黄体ホルモンを 10 分間隔で測定した時系列データ UKgas 1960 年 ~1986 年のイギリスのガス消費量を四半期ごとに観測した時系列データ ldeaths 1974 年 ~1979 年のイギリスで喘息 気管支炎 肺気腫による死亡数を月ごとに記録した時系列データ mdeaths
1 2 3 6 10 < > 13 16 16 4 17 13 00 15 30 5
2004 16 3 23 q 4 21 r 1 2 3 6 10 < > 13 16 16 4 17 13 00 15 30 5 13 2 2 16 4 4 17 3 16 3 1 16 3 2 905 1438 1201 1205 1210 70 1812 25 1635 1654 3 44 47 10 10 911.18-R 1193 34 1652 4 911.107-H 1159 1685
項 目
1 1 2 3 11 4 6 5 7,000 2 120 1.3 4,000 04 450 < > 5 3 6 7 8 9 4 10 11 5 12 45 6 13 E. 7 B. C. 14 15 16 17 18 19 20 21 22 23 8 24 25 9 27 2 26 6 27 3 1 3 3 28 29 30 9 31 32 33 500 1 4000 0 2~3 10 10 34
untitled
2 : n =1, 2,, 10000 0.5125 0.51 0.5075 0.505 0.5025 0.5 0.4975 0.495 0 2000 4000 6000 8000 10000 2 weak law of large numbers 1. X 1,X 2,,X n 2. µ = E(X i ),i=1, 2,,n 3. σi 2 = V (X i ) σ 2,i=1, 2,,n ɛ>0
1
1 5% 4% 11% 8% 13% 12% 10% 6% 17% 6% 8% 4% 6% 6% 2% 17% 17% 12% 14% 16% 6% 37% 11% 17% 35% 2 (N=6,239) 0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 1,585 1,126 950 494 345 296 242 263 191 150 131 116
RF_1
RF_1 10/04/16 10:32 http://rftechno.web.infoseek.co.jp/rf_1.html 1/12 RF_1 10/04/16 10:32 http://rftechno.web.infoseek.co.jp/rf_1.html 2/12 RF_1 10/04/16 10:32 http://rftechno.web.infoseek.co.jp/rf_1.html
●70974_100_AC009160_KAPヘ<3099>ーシス自動車約款（11.10）.indb
" # $ % & ' ( ) * +, -. / 0 1 2 3 4 5 6 7 8 9 : ; < = >? @ A B C D E F G H I J K L M N O P Q R S T U V W X Y " # $ % & ' ( ) * + , -. / 0 1 2 3 4 5 6 7 8 9 : ; < = > ? @ A B
08-Note2-web
r(t) t r(t) O v(t) = dr(t) dt a(t) = dv(t) dt = d2 r(t) dt 2 r(t), v(t), a(t) t dr(t) dt r(t) =(x(t),y(t),z(t)) = d 2 r(t) dt 2 = ( dx(t) dt ( d 2 x(t) dt 2, dy(t), dz(t) dt dt ), d2 y(t) dt 2, d2 z(t)
untitled
280 200 5 7,800 6 8,600 28 1 1 18 7 8 2 ( 31 ) 7 42 2 / / / / / / / / / / 1 3 (1) 4 5 3 1 1 1 A B C D 6 (1) -----) (2) -- ()) (3) ----(). ()() () ( )( )( )( ) ( ) ( )( )( )( ) () (). () ()() 7 () ( ) 1
ARspec_decomp.dvi
February 8, 0 auto-regresive mode AR ) AR.. t N fx t);x t);x3 t); ;xn t)g Burg AR xt) a m xt m t)+fft) AR M Fina Prediction Error,FPE) FPEM) ^ff M +MN MN ^ff M P f) P f) ff t M a m e ißfm t AR [,,3]. AR
中央大学理工学部情報工学科
04D8101033F 2008 3 1 1 2001 2006 1 1...1 2...2 2.1...2 2.1.1...2 2.1.2...3 2.2...4 2.2.1 AR...4 2.2.2 MA...5 2.2.3 ARMA...5 2.2.4 ARIMA...6 2.3...6 2.4...7 2.5...7 3...9 3.1... 11 3.2...13 3.3...16 3.4...18
untitled
ECOH/YG No. 6 556 754 On the Role o Spectral Width and Shape Parameters in Control o Wave Height Distribution Yoshimi GODA and Masanobu KUDAKA -4-6-4 ( -4-6-4 Distributions o individual wave heights approximately
keisoku01.dvi
2.,, Mon, 2006, 401, SAGA, JAPAN Dept. of Mechanical Engineering, Saga Univ., JAPAN 4 Mon, 2006, 401, SAGA, JAPAN Dept. of Mechanical Engineering, Saga Univ., JAPAN 5 Mon, 2006, 401, SAGA, JAPAN Dept.
x ( ) x dx = ax
x ( ) x dx = ax 1 dx = a x log x = at + c x(t) = e at C (C = e c ) a > 0 t a < 0 t 0 (at + b ) h dx = lim x(t + h) x(t) h 0 h x(t + h) x(t) h x(t) t x(t + h) x(t) ax(t) h x(t + h) x(t) + ahx(t) 0, h, 2h,
(u(x)v(x)) = u (x)v(x) + u(x)v (x) ( ) u(x) = u (x)v(x) u(x)v (x) v(x) v(x) 2 y = g(t), t = f(x) y = g(f(x)) dy dx dy dx = dy dt dt dx., y, f, g y = f (g(x))g (x). ( (f(g(x)). ). [ ] y = e ax+b (a, b )
McCain & McCleary (1979) The Statistical Analysis of the Simple Interrupted Time-Series Quasi-Experiment
Quasi-Experimenaion Ch.6 005/8/7 ypo rep: The Saisical Analysis of he Simple Inerruped Time-Series Quasi-Experimen INTRODUCTION () THE PROBLEM WITH ORDINAR LEAST SQUARE REGRESSION OLS (Ordinary Leas Square)
d (K + U) = v [ma F(r)] = (2.4.4) t = t r(t ) = r t 1 r(t 1 ) = r 1 U(r 1 ) U(r ) = t1 t du t1 = t F(r(t)) dr(t) r1 = F dr (2.4.5) r F 2 F ( F) r A r
![d (K + U) = v [ma F(r)] = (2.4.4) t = t r(t ) = r t 1 r(t 1 ) = r 1 U(r 1 ) U(r ) = t1 t du t1 = t F(r(t)) dr(t) r1 = F dr (2.4.5) r F 2 F ( F) r A r](/thumbs/85/92098187.jpg "d (K + U) = v [ma F(r)] = (2.4.4) t = t r(t ) = r t 1 r(t 1 ) = r 1 U(r 1 ) U(r ) = t1 t du t1 = t F(r(t)) dr(t) r1 = F dr (2.4.5) r F 2 F ( F) r A r")
2.4 ( ) U(r) ( ) ( ) U F(r) = x, U y, U = U(r) (2.4.1) z 2 1 K = mv 2 /2 dk = d ( ) 1 2 mv2 = mv dv = v (ma) (2.4.2) ( ) U(r(t)) r(t) r(t) + dr(t) du du = U(r(t) + dr(t)) U(r(t)) = U x = U(r(t)) dr(t)
-- o C inter Arctic Oscillation Index Sapporo Air-Temp Matlab randn (red n
--.........................4..3. (autoregressive process)...5..4....6..5. (MA)...8..6. (ARMA)...8..7. (ARIMA)...8..8....8.3....8.3......4....3...4.. - (persistency) (serial correlation) m amias (988) 970
k2 ( :35 ) ( k2) (GLM) web web 1 :
2012 11 01 k2 (2012-10-26 16:35 ) 1 6 2 (2012 11 01 k2) (GLM) kubo@ees.hokudai.ac.jp web http://goo.gl/wijx2 web http://goo.gl/ufq2 1 : 2 2 4 3 7 4 9 5 : 11 5.1................... 13 6 14 6.1......................
<4D F736F F D20939D8C7689F090CD985F93C18EEA8D758B E646F63>
Gretl OLS omitted variable omitted variable AIC,BIC a) gretl gretl sample file Greene greene8_3 Add Define new variable l_g_percapita=log(g/pop) Pg,Y,Pnc,Puc,Ppt,Pd,Pn,Ps Add logs of selected variables
「国債の金利推定モデルに関する研究会」報告書
: LG 19 7 26 2 LG Quadratic Gaussian 1 30 30 3 4 2,,, E-mail: kijima@center.tmu.ac.jp, E-mail: tanaka-keiichi@tmu.ac.jp 1 L G 2 1 L G r L t),r G t) L r L t) G r G t) r L t) h G t) =r G t) r L t) r L t)
SML-01 series : LED
Data Sheet SML- series 3228 3.2.mmt=.3mm 3.2.mmt=.3mm Color Type V U D W M P E B WB PD(mW) IF(mA) (Ta=25) IFP(mA) VR(V) Topr() Tstg() (Ta=25) VF IR D IV Typ. (V) IF(mA) (A) Max. Min*2 VR(V) Typ. (nm) (nm)
F0 P( T, K) C ( TK, ) exp ( rt) < dk + 3 dk F K K # # 0 r K T P C S 0 0 F0= exp ( rt) S0T 0 F0
日 経 5 株 価 指 数 のモデル フリー インプライド ボラ Titleティリティの 計 算 方 法 に 関 して : ボラティリティ 予 測 力 の 観 点 から Author(s) 山 口, 圭 子 Citation 一 橋 経 済 学, 3(1): 9-43 Issue 008-07-0 Date Type Departmental Bulletin Paper Text Version
dpri04.dvi
47 B 6 4 Annuals of Disas. Prev. Res. Inst., Kyoto Univ., No. 47B, 24 AR AR :,, AR,. () point process Waymire et et al. (984) WGR () Over and Gupta (994, 996) Chatchai et al. (2) (23) Fig. Structure for
2
1 2 119 119 5 500 1 30 102 1 113 3 4 120 2 3 113 5 230 1 1 3 4 5 6 7 8 1 support@kansen.sakura.ne.jp 2 9 3 ( ) 10 11 12 4 1. 2. 3. 4. 13 5 14 15 16 17 18 19 [ ] [ ] 20 [ ] [ ] [ ] 21 22 [ ] 23 < > < >
KENZOU
KENZOU 2008 8 2 3 2 3 2 2 4 2 4............................................... 2 4.2............................... 3 4.2........................................... 4 4.3..............................
1 1 2 65
3 3 2000 6 14 2 30 4 2 1 1 2 65 1!?? < > 3 2 2 100 19 19 100 100 100 < > 19 2 2 2 2 < > 2000 2000 50 1945 5 50 1945 5 45 20 20 4 1945 4 5 5 5 100 50 20 5 2 20 5 20 5 5 6 20 6 19 5 5 6 5 6 2 20 6 21
一般化線形 (混合) モデル (2) - ロジスティック回帰と GLMM
.. ( ) (2) GLMM kubo@ees.hokudai.ac.jp I http://goo.gl/rrhzey 2013 08 27 : 2013 08 27 08:29 kubostat2013ou2 (http://goo.gl/rrhzey) ( ) (2) 2013 08 27 1 / 74 I.1 N k.2 binomial distribution logit link function.3.4!
~こ 分 ~ネ( / ) グループ (II) 1は 表 1 の 20, 21 の 疑 問 文 すなわち 文 末 形 式 が r~ ショウカ( 意 志 ) 場 合 が 付 く 場 合 の 表 現 機 能 の 違 いが 顕 著 である まずは Y/N~ショウカ 1 {~ね> 付 加 の 文 とも 付 加 の 文 とも 異 なる (ll 語 文 ~ r 情 報 論 における W~
laplace.dvi
Λ 2.1 2004.2.20 1 Λ 1 2 Ay = u 2 2 A 2 u " # a 11 a 12 A = ; u = a 21 a 22 " # u 1 u 2 y Ay = u (1) A (1) y = A 1 u y A 2 x i i i =1; 2 Ax 1 = 1 x 1 ; Ax 2 = 2 x 2 (2) x 1 x 2 =0 (3) (3) (2) x 1 x 2 x
CDMA (high-compaciton multicarrier codedivision multiple access: HC/MC-CDMA),., HC/MC-CDMA,., 32.,, 64. HC/MC-CDMA, HC-MCM, i
24 Investigation on HC/MC-CDMA Signals with Non-Uniform Frequency Intervals 1130401 2013 3 1 CDMA (high-compaciton multicarrier codedivision multiple access: HC/MC-CDMA),., HC/MC-CDMA,., 32.,, 64. HC/MC-CDMA,
1 7 ω ω ω 7.1 0, ( ) Q, 7.2 ( Q ) 7.1 ω Z = R +jx Z 1/ Z 7.2 ω 7.2 Abs. admittance (x10-3 S) RLC Series Circuit Y R = 20 Ω L = 100
7 7., ) Q, 7. Q ) 7. Z = R +jx Z / Z 7. 7. Abs. admittance x -3 S) 5 4 3 R Series ircuit Y R = Ω = mh = uf Q = 5 5 5 V) Z = R + jx 7. Z 7. ) R = Ω = mh = µf ) 7 V) R Z s = R + j ) 7.3 R =. 7.4) ) f = π.
1 5 1.1..................................... 5 1.2..................................... 5 1.3.................................... 6 2 OSPF 7 2.1 OSPF.
2011 2012 1 31 5110B036-6 1 5 1.1..................................... 5 1.2..................................... 5 1.3.................................... 6 2 OSPF 7 2.1 OSPF....................................
1 1 ( ) ( 1.1 1.1.1 60% mm 100 100 60 60% 1.1.2 A B A B A 1
1 21 10 5 1 E-mail: qliu@res.otaru-uc.ac.jp 1 1 ( ) ( 1.1 1.1.1 60% mm 100 100 60 60% 1.1.2 A B A B A 1 B 1.1.3 boy W ID 1 2 3 DI DII DIII OL OL 1.1.4 2 1.1.5 1.1.6 1.1.7 1.1.8 1.2 1.2.1 1. 2. 3 1.2.2
山形大学紀要
x t IID t = b b x t t x t t = b t- AR ARMA IID AR ARMAMA TAR ARCHGARCH TARThreshold Auto Regressive Model TARTongTongLim y y X t y Self Exciting Threshold Auto Regressive, SETAR SETARTAR TsayGewekeTerui
1 8, : 8.1 1, 2 z = ax + by + c ax by + z c = a b +1 x y z c = 0, (0, 0, c), n = ( a, b, 1). f = n i=1 a ii x 2 i + i<j 2a ij x i x j = ( x, A x), f =
1 8, : 8.1 1, z = ax + by + c ax by + z c = a b +1 x y z c = 0, (0, 0, c), n = ( a, b, 1). f = a ii x i + i
36
36 37 38 P r R P 39 (1+r ) P =R+P g P r g P = R r g r g == == 40 41 42 τ R P = r g+τ 43 τ (1+r ) P τ ( P P ) = R+P τ ( P P ) n P P r P P g P 44 R τ P P = (1 τ )(r g) (1 τ )P R τ 45 R R σ u R= R +u u~ (0,σ
Part 1 GARCH () ( ) /24, p.2/93
基盤研究 A 統計科学における数理的手法の理論と応用 ( 研究代表者 : 谷口正信 ) によるシンポジウム 計量ファイナンスと時系列解析法の新たな展開 平成 20 年 1 月 24 日 ~26 日香川大学 Realized Volatility の長期記憶性について 1 研究代表者 : 前川功一 ( 広島経済大学 ) 共同研究者 : 得津康義 ( 広島経済大学 ) 河合研一 ( 統計数理研究所リスク解析戦略研究センター
IS-LM (interest) 100 (net rate of interest) (rate of interest) ( ) = 100 (2.1) (gross rate of interest) ( ) = 100 (2.2)
1 2 2 2 2.1 IS-LM 1 2.2 1 1 (interest) 100 (net rate of interest) (rate of interest) ( ) = 100 (2.1) (gross rate of interest) ( ) = 100 (2.2) 1 1. 2. 1 1 ( ) 2.3. 3 2.3 1 (yield to maturity) (rate of return)
AR(1) y t = φy t 1 + ɛ t, ɛ t N(0, σ 2 ) 1. Mean of y t given y t 1, y t 2, E(y t y t 1, y t 2, ) = φy t 1 2. Variance of y t given y t 1, y t
87 6.1 AR(1) y t = φy t 1 + ɛ t, ɛ t N(0, σ 2 ) 1. Mean of y t given y t 1, y t 2, E(y t y t 1, y t 2, ) = φy t 1 2. Variance of y t given y t 1, y t 2, V(y t y t 1, y t 2, ) = σ 2 3. Thus, y t y t 1,
214 March 31, 214, Rev.2.1 4........................ 4........................ 5............................. 7............................... 7 1 8 1.1............................... 8 1.2.......................
213 March 25, 213, Rev.1.5 4........................ 4........................ 6 1 8 1.1............................... 8 1.2....................... 9 2 14 2.1..................... 14 2.2............................
tokei01.dvi
2. :,,,. :.... Apr. - Jul., 26FY Dept. of Mechanical Engineering, Saga Univ., JAPAN 4 3. (probability),, 1. : : n, α A, A a/n. :, p, p Apr. - Jul., 26FY Dept. of Mechanical Engineering, Saga Univ., JAPAN
0 = m 2p 1 p = 1/2 p y = 1 m = 1 2 d ( + 1)2 d ( + 1) 2 = d d ( + 1)2 = = 2( + 1) 2 g() 2 f() f() = [g()] 2 = g()g() f f () = [g()g()]
![0 = m 2p 1 p = 1/2 p y = 1 m = 1 2 d ( + 1)2 d ( + 1) 2 = d d ( + 1)2 = = 2( + 1) 2 g() 2 f() f() = [g()] 2 = g()g() f f () = [g()g()]](/thumbs/92/110082312.jpg "0 = m 2p 1 p = 1/2 p y = 1 m = 1 2 d ( + 1)2 d ( + 1) 2 = d d ( + 1)2 = = 2( + 1) 2 g() 2 f() f() = [g()] 2 = g()g() f f () = [g()g()]")
8. 2 1 2 1 2 ma,y u(, y) s.t. p + p y y = m u y y p p y y m u(, y) = y p + p y y = m y ( ) 1 y = (m p ) p y = m p y p p y 2 0 m/p U U() = m p y p p y 2 2 du() d = m p y 2p p y 1 0 = m 2p 1 p = 1/2 p y
untitled
MCMC 2004 23 1 I. MCMC 1. 2. 3. 4. MH 5. 6. MCMC 2 II. 1. 2. 3. 4. 5. 3 I. MCMC 1. 2. 3. 4. MH 5. 4 1. MCMC 5 2. A P (A) : P (A)=0.02 A B A B Pr B A) Pr B A c Pr B A)=0.8, Pr B A c =0.1 6 B A 7 8 A, :
講義のーと : データ解析のための統計モデリング. 第２回
Title 講義のーと : データ解析のための統計モデリング Author(s) 久保, 拓弥 Issue Date 2008 Doc URL http://hdl.handle.net/2115/49477 Type learningobject Note この講義資料は, 著者のホームページ http://hosho.ees.hokudai.ac.jp/~kub ードできます Note(URL)http://hosho.ees.hokudai.ac.jp/~kubo/ce/EesLecture20
(time series) ( 225 ) / / p.2/66
338 857 255 Tel : 48 858 3577, Fax : 48 858 3716 Email : tohru@ics.saitama-u.ac.jp URL : http://www.nls.ics.saitama-u.ac.jp/ tohru / / p.1/66 (time series) ( 225 ) / / p.2/66 / / p.3/66 ?? / / p.3/66 1.9.8.7.6???.5.4.3.2.1
ii
i 2013 5 143 5.1...................................... 143 5.2.................................. 144 5.3....................................... 148 5.4.................................. 153 5.5...................................
, 183
182 , 183 < > < > 184 < > < > < p. > 185 < > < > 186 < p. > < p. > < p. > < p. > < p. > < p. > < p. > 187 < p. > 188 < p. > 189 190 A B A B A B A
DE-resume
- 2011, http://c-faculty.chuo-u.ac.jp/ nishioka/ 2 11 21131 : 4 1 x y(x, y (x,y (x,,y (n, (1.1 F (x, y, y,y,,y (n =0. (1.1 n. (1.1 y(x. y(x (1.1. 1 1 1 1.1... 2 1.2... 9 1.3 1... 26 2 2 34 2.1,... 35 2.2
3 10 14 17 25 30 35 43 2
THE ASSOCIATION FOR REAL ESTATE SECURITIZATION 40 2009 July-August 3 10 14 17 25 30 35 43 2 ARES SPECIAL ARES July-August 2009 3 4 ARES July-August 2009 ARES SPECIAL 5 ARES July-August 2009 ARES SPECIAL
211 kotaro@math.titech.ac.jp 1 R *1 n n R n *2 R n = {(x 1,..., x n ) x 1,..., x n R}. R R 2 R 3 R n R n R n D D R n *3 ) (x 1,..., x n ) f(x 1,..., x n ) f D *4 n 2 n = 1 ( ) 1 f D R n f : D R 1.1. (x,