library(rlang)
library('wavelets')
library('readxl')
library('ggplot2')
library('dplyr')
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
df <- read_excel("C:/Users/LENOVO/Downloads/Data Laprak 3 ARW.xlsx")
data <- df$Penumpang
data
## [1] 11504 11705 13066 12657 13312 12929 14087 12991 13183 13949 13369 13344
## [13] 13647 10759 13167 14189 14939 14820 16092 15144 14792 15465 14141 14766
## [25] 14697 14116 15756 15435 16054 16636 17463 16769 15604 16901 15696 15011
## [37] 14246 13593 16826 16458 17467 17746 17959 17204 16840 17438 16434 17211
## [49] 17039 14880 16617 16156 16565 16808 17181 16140 16713 16542 16088 17299
## [61] 16446 14519 16584 16031 17018 16806 17632 14492 16353 16062 15765 16333
## [73] 15801 15126 16701 16376 17401 17687 17956 16675 16063 16828 15436 15745
## [85] 14573 14315 15521 15724 15795 16932 19917 19031 19439 20198 19578 20992
## [97] 20698 19628 22427 21502 22547 23415 22125 22763 23219 24503 23986 25791
## [109] 24254 22394 26841 26150 27450 27118 27077 27351 27125 28280 27253 29328
## [121] 27886 26058 28156 28000 30176 28730 28216 29125 29019 29765 29178 31530
## [133] 30359 26837 31612 30934 33157 30181 33669 33255 31921 34498 33798 36140
## [145] 34107 30721 35272 35135 34877 32270 36089 34560 33878 35602 34637 37197
## [157] 34435 31282 35068 35106 34514 34261 38303 34542 34615 35814 35228 36710
# j4 aj1 before
modwt_timeseries <- function (x,w='haar',j=4) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
v4=x.modwt@V$V4
x1<-x2<-x3<-x4<-x5<-NULL
for (i in 1:(n-17)){
x1<-c(x1,d1[i+16])
x2<-c(x2,d2[i+16])
x3<-c(x3,d3[i+16])
x4<-c(x4,d4[i+16])
x5<-c(x5,v4[i+16])
}
y=x[18:n]
lm.y=lm(y~-1+x1+x2+x3+x4+x5)
koef<-lm.y$coeff
pred<-c(rep(0,17), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4610.9 -701.6 105.1 895.5 3589.5
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.08914 0.13917 0.641 0.523
## x2 0.27084 0.18862 1.436 0.153
## x3 0.85414 0.21281 4.014 9.52e-05 ***
## x4 1.00226 0.18467 5.427 2.33e-07 ***
## x5 1.01579 0.00686 148.074 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1356 on 146 degrees of freedom
## Multiple R-squared: 0.9969, Adjusted R-squared: 0.9968
## F-statistic: 9518 on 5 and 146 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5)
##
## Coefficients:
## x1 x2 x3 x4 x5
## 0.08914 0.27084 0.85414 1.00227 1.01579
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 17), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4610.9 -701.6 105.1 895.5 3589.5
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.08914 0.13917 0.641 0.523
## x2 0.27084 0.18862 1.436 0.153
## x3 0.85414 0.21281 4.014 9.52e-05 ***
## x4 1.00226 0.18467 5.427 2.33e-07 ***
## x5 1.01579 0.00686 148.074 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1356 on 146 degrees of freedom
## Multiple R-squared: 0.9969, Adjusted R-squared: 0.9968
## F-statistic: 9518 on 5 and 146 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
# j4 aj1 after
modwt_timeseries <- function (x,w='haar',j=4) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
v4=x.modwt@V$V4
x1<-x2<-x3<-x4<-x5<-NULL
for (i in 1:(n-17)){
x1<-c(x1,d1[i+16])
x2<-c(x2,d2[i+16])
x3<-c(x3,d3[i+16])
x4<-c(x4,d4[i+16])
x5<-c(x5,v4[i+16])
}
y=x[18:n]
lm.y=lm(y~-1+x3+x4+x5)
koef<-lm.y$coeff
pred<-c(rep(0,17), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x3 + x4 + x5)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4715.9 -643.9 62.9 1006.3 3699.1
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x3 0.920944 0.208751 4.412 1.96e-05 ***
## x4 1.010520 0.184870 5.466 1.91e-07 ***
## x5 1.016705 0.006846 148.515 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1359 on 148 degrees of freedom
## Multiple R-squared: 0.9969, Adjusted R-squared: 0.9968
## F-statistic: 1.581e+04 on 3 and 148 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x3 + x4 + x5)
##
## Coefficients:
## x3 x4 x5
## 0.9209 1.0105 1.0167
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 17), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x3 + x4 + x5)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4715.9 -643.9 62.9 1006.3 3699.1
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x3 0.920944 0.208751 4.412 1.96e-05 ***
## x4 1.010520 0.184870 5.466 1.91e-07 ***
## x5 1.016705 0.006846 148.515 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1359 on 148 degrees of freedom
## Multiple R-squared: 0.9969, Adjusted R-squared: 0.9968
## F-statistic: 1.581e+04 on 3 and 148 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
# j4 aj2 before
modwt_timeseries <- function (x,w='haar',j=4) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
v4=x.modwt@V$V4
x1<-x2<-x3<-x4<-x5<-x6<-x7<-x8<-x9<-x10<-NULL
for (i in 1:(n-17)){
x1<-c(x1,d1[i+16])
x2<-c(x2,d1[i+14])
x3<-c(x3,d2[i+16])
x4<-c(x4,d2[i+12])
x5<-c(x5,d3[i+16])
x6<-c(x6,d3[i+8])
x7<-c(x7,d4[i+16])
x8<-c(x8,d4[i])
x9<-c(x9,v4[i+16])
x10<-c(x10,v4[i])
}
y=x[18:n]
lm.y=lm(y~-1+x1+x2+x3+x4+x5+x6+x7+x8+x9+x10)
koef<-lm.y$coeff
pred<-c(rep(0,17), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 +
## x9 + x10)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3963.4 -615.4 -30.4 691.6 3427.8
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.10018 0.13143 0.762 0.447190
## x2 0.16941 0.13134 1.290 0.199233
## x3 0.20490 0.17900 1.145 0.254283
## x4 -0.10849 0.18581 -0.584 0.560219
## x5 1.15227 0.21259 5.420 2.51e-07 ***
## x6 0.97363 0.21648 4.498 1.42e-05 ***
## x7 0.75613 0.20392 3.708 0.000299 ***
## x8 0.05251 0.07375 0.712 0.477641
## x9 0.97197 0.03799 25.588 < 2e-16 ***
## x10 0.03568 0.03822 0.933 0.352237
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1274 on 141 degrees of freedom
## Multiple R-squared: 0.9974, Adjusted R-squared: 0.9972
## F-statistic: 5392 on 10 and 141 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 +
## x9 + x10)
##
## Coefficients:
## x1 x2 x3 x4 x5 x6 x7 x8
## 0.10018 0.16941 0.20490 -0.10849 1.15227 0.97363 0.75613 0.05251
## x9 x10
## 0.97197 0.03568
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 17), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 +
## x9 + x10)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3963.4 -615.4 -30.4 691.6 3427.8
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.10018 0.13143 0.762 0.447190
## x2 0.16941 0.13134 1.290 0.199233
## x3 0.20490 0.17900 1.145 0.254283
## x4 -0.10849 0.18581 -0.584 0.560219
## x5 1.15227 0.21259 5.420 2.51e-07 ***
## x6 0.97363 0.21648 4.498 1.42e-05 ***
## x7 0.75613 0.20392 3.708 0.000299 ***
## x8 0.05251 0.07375 0.712 0.477641
## x9 0.97197 0.03799 25.588 < 2e-16 ***
## x10 0.03568 0.03822 0.933 0.352237
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1274 on 141 degrees of freedom
## Multiple R-squared: 0.9974, Adjusted R-squared: 0.9972
## F-statistic: 5392 on 10 and 141 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
# j4 aj2 after
modwt_timeseries <- function (x,w='haar',j=4) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
v4=x.modwt@V$V4
x1<-x2<-x3<-x4<-x5<-x6<-x7<-x8<-x9<-x10<-NULL
for (i in 1:(n-17)){
x1<-c(x1,d1[i+16])
x2<-c(x2,d1[i+14])
x3<-c(x3,d2[i+16])
x4<-c(x4,d2[i+12])
x5<-c(x5,d3[i+16])
x6<-c(x6,d3[i+8])
x7<-c(x7,d4[i+16])
x8<-c(x8,d4[i])
x9<-c(x9,v4[i+16])
x10<-c(x10,v4[i])
}
y=x[18:n]
lm.y=lm(y~-1+x5+x6+x7+x9)
koef<-lm.y$coeff
pred<-c(rep(0,17), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x5 + x6 + x7 + x9)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4192.1 -671.4 21.6 788.9 3536.2
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x5 1.226033 0.206003 5.952 1.87e-08 ***
## x6 0.997206 0.213116 4.679 6.48e-06 ***
## x7 0.677824 0.187094 3.623 0.000401 ***
## x9 1.007587 0.006698 150.429 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1272 on 147 degrees of freedom
## Multiple R-squared: 0.9973, Adjusted R-squared: 0.9972
## F-statistic: 1.354e+04 on 4 and 147 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x5 + x6 + x7 + x9)
##
## Coefficients:
## x5 x6 x7 x9
## 1.2260 0.9972 0.6778 1.0076
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 17), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x5 + x6 + x7 + x9)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4192.1 -671.4 21.6 788.9 3536.2
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x5 1.226033 0.206003 5.952 1.87e-08 ***
## x6 0.997206 0.213116 4.679 6.48e-06 ***
## x7 0.677824 0.187094 3.623 0.000401 ***
## x9 1.007587 0.006698 150.429 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1272 on 147 degrees of freedom
## Multiple R-squared: 0.9973, Adjusted R-squared: 0.9972
## F-statistic: 1.354e+04 on 4 and 147 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
# j5 aj1 before
modwt_timeseries <- function (x,w='haar',j=5) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
d5=x.modwt@W$W5
v5=x.modwt@V$V5
x1<-x2<-x3<-x4<-x5<-x6<-NULL
for (i in 1:(n-33)){
x1<-c(x1,d1[i+32])
x2<-c(x2,d2[i+32])
x3<-c(x3,d3[i+32])
x4<-c(x4,d4[i+32])
x5<-c(x5,d5[i+32])
x6<-c(x6,v5[i+32])
}
y=x[34:n]
lm.y=lm(y~-1+x1+x2+x3+x4+x5+x6)
koef<-lm.y$coeff
pred<-c(rep(0,33), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4652.9 -747.5 95.4 839.2 3664.1
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.078220 0.146313 0.535 0.593843
## x2 0.204330 0.201529 1.014 0.312528
## x3 0.853323 0.238098 3.584 0.000479 ***
## x4 0.943103 0.232272 4.060 8.45e-05 ***
## x5 1.137435 0.172344 6.600 9.69e-10 ***
## x6 1.010127 0.009692 104.220 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1406 on 129 degrees of freedom
## Multiple R-squared: 0.997, Adjusted R-squared: 0.9968
## F-statistic: 7059 on 6 and 129 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6)
##
## Coefficients:
## x1 x2 x3 x4 x5 x6
## 0.07822 0.20433 0.85332 0.94310 1.13743 1.01013
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 33), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4652.9 -747.5 95.4 839.2 3664.1
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.078220 0.146313 0.535 0.593843
## x2 0.204330 0.201529 1.014 0.312528
## x3 0.853323 0.238098 3.584 0.000479 ***
## x4 0.943103 0.232272 4.060 8.45e-05 ***
## x5 1.137435 0.172344 6.600 9.69e-10 ***
## x6 1.010127 0.009692 104.220 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1406 on 129 degrees of freedom
## Multiple R-squared: 0.997, Adjusted R-squared: 0.9968
## F-statistic: 7059 on 6 and 129 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
# j5 aj1 after
modwt_timeseries <- function (x,w='haar',j=5) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
d5=x.modwt@W$W5
v5=x.modwt@V$V5
x1<-x2<-x3<-x4<-x5<-x6<-NULL
for (i in 1:(n-33)){
x1<-c(x1,d1[i+32])
x2<-c(x2,d2[i+32])
x3<-c(x3,d3[i+32])
x4<-c(x4,d4[i+32])
x5<-c(x5,d5[i+32])
x6<-c(x6,v5[i+32])
}
y=x[34:n]
lm.y=lm(y~-1+x3+x4+x5+x6)
koef<-lm.y$coeff
pred<-c(rep(0,33), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x3 + x4 + x5 + x6)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4744.3 -716.0 17.1 883.0 3758.8
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x3 0.902901 0.233323 3.870 0.000171 ***
## x4 0.951328 0.231523 4.109 6.96e-05 ***
## x5 1.140659 0.171856 6.637 7.70e-10 ***
## x6 1.010570 0.009659 104.622 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1402 on 131 degrees of freedom
## Multiple R-squared: 0.9969, Adjusted R-squared: 0.9968
## F-statistic: 1.064e+04 on 4 and 131 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x3 + x4 + x5 + x6)
##
## Coefficients:
## x3 x4 x5 x6
## 0.9029 0.9513 1.1407 1.0106
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 33), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x3 + x4 + x5 + x6)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4744.3 -716.0 17.1 883.0 3758.8
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x3 0.902901 0.233323 3.870 0.000171 ***
## x4 0.951328 0.231523 4.109 6.96e-05 ***
## x5 1.140659 0.171856 6.637 7.70e-10 ***
## x6 1.010570 0.009659 104.622 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1402 on 131 degrees of freedom
## Multiple R-squared: 0.9969, Adjusted R-squared: 0.9968
## F-statistic: 1.064e+04 on 4 and 131 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
# j5 aj2 before
modwt_timeseries <- function (x,w='haar',j=5) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
d5=x.modwt@W$W5
v5=x.modwt@V$V5
x1<-x2<-x3<-x4<-x5<-x6<-x7<-x8<-x9<-x10 <- x11 <- x12<-NULL
for (i in 1:(n-33)){
x1<-c(x1,d1[i+32])
x2<-c(x2,d1[i+30])
x3<-c(x3,d2[i+32])
x4<-c(x4,d2[i+28])
x5<-c(x5,d3[i+32])
x6<-c(x6,d3[i+24])
x7<-c(x7,d4[i+32])
x8<-c(x8,d4[i+16])
x9<-c(x9,d5[i+32])
x10<-c(x10,d5[i])
x11<-c(x11,v5[i+32])
x12<-c(x12,v5[i])
}
y=x[34:n]
lm.y=lm(y~-1+x1+x2+x3+x4+x5+x6+x7+x8+x9+x10+x11+x12)
koef<-lm.y$coeff
pred<-c(rep(0,33), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 +
## x9 + x10 + x11 + x12)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3996.5 -539.3 18.6 681.6 3449.0
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.104178 0.140015 0.744 0.458268
## x2 0.155851 0.139973 1.113 0.267690
## x3 0.200672 0.192753 1.041 0.299878
## x4 -0.101963 0.203299 -0.502 0.616886
## x5 1.296472 0.253750 5.109 1.2e-06 ***
## x6 1.025869 0.261367 3.925 0.000143 ***
## x7 0.825482 0.239412 3.448 0.000774 ***
## x8 0.285650 0.239050 1.195 0.234410
## x9 0.744673 0.210185 3.543 0.000560 ***
## x10 0.029890 0.052964 0.564 0.573547
## x11 0.999291 0.027677 36.106 < 2e-16 ***
## x12 0.009052 0.026930 0.336 0.737348
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1339 on 123 degrees of freedom
## Multiple R-squared: 0.9974, Adjusted R-squared: 0.9971
## F-statistic: 3894 on 12 and 123 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 +
## x9 + x10 + x11 + x12)
##
## Coefficients:
## x1 x2 x3 x4 x5 x6 x7
## 0.104178 0.155851 0.200672 -0.101963 1.296472 1.025869 0.825482
## x8 x9 x10 x11 x12
## 0.285650 0.744673 0.029890 0.999291 0.009052
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 33), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 +
## x9 + x10 + x11 + x12)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3996.5 -539.3 18.6 681.6 3449.0
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x1 0.104178 0.140015 0.744 0.458268
## x2 0.155851 0.139973 1.113 0.267690
## x3 0.200672 0.192753 1.041 0.299878
## x4 -0.101963 0.203299 -0.502 0.616886
## x5 1.296472 0.253750 5.109 1.2e-06 ***
## x6 1.025869 0.261367 3.925 0.000143 ***
## x7 0.825482 0.239412 3.448 0.000774 ***
## x8 0.285650 0.239050 1.195 0.234410
## x9 0.744673 0.210185 3.543 0.000560 ***
## x10 0.029890 0.052964 0.564 0.573547
## x11 0.999291 0.027677 36.106 < 2e-16 ***
## x12 0.009052 0.026930 0.336 0.737348
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1339 on 123 degrees of freedom
## Multiple R-squared: 0.9974, Adjusted R-squared: 0.9971
## F-statistic: 3894 on 12 and 123 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
# j5 aj2 after
modwt_timeseries <- function (x,w='haar',j=5) {
n=length(x)
x.modwt=modwt(x,w,j)
d1=x.modwt@W$W1
d2=x.modwt@W$W2
d3=x.modwt@W$W3
d4=x.modwt@W$W4
d5=x.modwt@W$W5
v5=x.modwt@V$V5
x1<-x2<-x3<-x4<-x5<-x6<-x7<-x8<-x9<-x10 <- x11 <- x12<-NULL
for (i in 1:(n-33)){
x1<-c(x1,d1[i+32])
x2<-c(x2,d1[i+30])
x3<-c(x3,d2[i+32])
x4<-c(x4,d2[i+28])
x5<-c(x5,d3[i+32])
x6<-c(x6,d3[i+24])
x7<-c(x7,d4[i+32])
x8<-c(x8,d4[i+16])
x9<-c(x9,d5[i+32])
x10<-c(x10,d5[i])
x11<-c(x11,v5[i+32])
x12<-c(x12,v5[i])
}
y=x[34:n]
lm.y=lm(y~-1+x5+x6+x7+x9+x11)
koef<-lm.y$coeff
pred<-c(rep(0,33), lm.y$fitted)
print(summary(lm.y))
return(lm.y)
}
modwt_timeseries(data)
##
## Call:
## lm(formula = y ~ -1 + x5 + x6 + x7 + x9 + x11)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4230.9 -654.9 58.9 826.9 3496.2
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x5 1.305798 0.242280 5.390 3.21e-07 ***
## x6 1.042135 0.258080 4.038 9.16e-05 ***
## x7 0.756270 0.224340 3.371 0.000986 ***
## x9 0.868622 0.176021 4.935 2.41e-06 ***
## x11 1.011774 0.009145 110.638 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1327 on 130 degrees of freedom
## Multiple R-squared: 0.9973, Adjusted R-squared: 0.9972
## F-statistic: 9513 on 5 and 130 DF, p-value: < 2.2e-16
##
## Call:
## lm(formula = y ~ -1 + x5 + x6 + x7 + x9 + x11)
##
## Coefficients:
## x5 x6 x7 x9 x11
## 1.3058 1.0421 0.7563 0.8686 1.0118
data_plot <- data.frame(
n = c(1:length(data)),
Data = data,
Prediksi = c(rep(0, 33), modwt_timeseries(data)$fitted)
)
##
## Call:
## lm(formula = y ~ -1 + x5 + x6 + x7 + x9 + x11)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4230.9 -654.9 58.9 826.9 3496.2
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## x5 1.305798 0.242280 5.390 3.21e-07 ***
## x6 1.042135 0.258080 4.038 9.16e-05 ***
## x7 0.756270 0.224340 3.371 0.000986 ***
## x9 0.868622 0.176021 4.935 2.41e-06 ***
## x11 1.011774 0.009145 110.638 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1327 on 130 degrees of freedom
## Multiple R-squared: 0.9973, Adjusted R-squared: 0.9972
## F-statistic: 9513 on 5 and 130 DF, p-value: < 2.2e-16
data_plot %>% ggplot(aes(x=n)) +
geom_line(aes(y=Data), color='black') +
geom_line(aes(y=Prediksi), color='red') +
labs(title = "Prediksi vs Data Asli")
