#memanggil data dan library
data=read.table(file.choose(), header=T)
data
## Y X1 X2 X3
## 1 57.5 78 2.75 29.5
## 2 52.8 69 2.15 26.3
## 3 61.3 77 4.41 32.2
## 4 67.0 88 5.52 36.5
## 5 53.5 67 3.21 27.2
## 6 62.7 80 4.32 27.7
## 7 56.2 74 2.31 28.3
## 8 68.5 94 4.30 30.3
## 9 69.2 102 3.71 28.7
library(zoo)
## Warning: package 'zoo' was built under R version 4.5.3
##
## Attaching package: 'zoo'
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
library(car)
## Loading required package: carData
library(nortest)
library(lmtest)
## Warning: package 'lmtest' was built under R version 4.5.3
#Model 1
M1=lm(formula=Y~X1, data=data)
summary(M1)
##
## Call:
## lm(formula = Y ~ X1, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.6440 -1.9128 -0.2151 2.2513 2.4075
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 19.01108 5.42272 3.506 0.009915 **
## X1 0.51797 0.06635 7.807 0.000106 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.174 on 7 degrees of freedom
## Multiple R-squared: 0.897, Adjusted R-squared: 0.8823
## F-statistic: 60.95 on 1 and 7 DF, p-value: 0.0001065
#Pengujian Asumsi
#uji normalitas residual X1
shapiro.test(resid(M1))
##
## Shapiro-Wilk normality test
##
## data: resid(M1)
## W = 0.87986, p-value = 0.1565
#uji non-heteroskedastisitas X1
bptest(M1,varformula=~fitted.values(M1), studentize=F)
##
## Breusch-Pagan test
##
## data: M1
## BP = 0.35178, df = 1, p-value = 0.5531
#uji autokorelasi X1
dwtest(M1, alternative='two.sided')
##
## Durbin-Watson test
##
## data: M1
## DW = 1.785, p-value = 0.5744
## alternative hypothesis: true autocorrelation is not 0
#uji autokorelasi X1
bgtest(M1)
##
## Breusch-Godfrey test for serial correlation of order up to 1
##
## data: M1
## LM test = 0.033549, df = 1, p-value = 0.8547
#uji glejser
u1=resid(M1)
u1abs=abs(u1)
M1_abs=lm(formula=u1abs~X1, data=data)
summary(M1_abs)
##
## Call:
## lm(formula = u1abs ~ X1, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.3200 -0.4062 0.2952 0.5324 0.7722
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.57193 2.04979 -0.279 0.788
## X1 0.02864 0.02508 1.142 0.291
##
## Residual standard error: 0.8219 on 7 degrees of freedom
## Multiple R-squared: 0.157, Adjusted R-squared: 0.03657
## F-statistic: 1.304 on 1 and 7 DF, p-value: 0.2911
#Model 2
M2=lm(formula=Y~X2, data=data)
summary(M2)
##
## Call:
## lm(formula = Y ~ X2, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.6496 -2.1172 -1.2392 0.9339 7.8929
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 45.298 5.254 8.621 5.64e-05 ***
## X2 4.315 1.390 3.105 0.0172 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.394 on 7 degrees of freedom
## Multiple R-squared: 0.5793, Adjusted R-squared: 0.5192
## F-statistic: 9.64 on 1 and 7 DF, p-value: 0.0172
#Pengujian Asumsi
#uji normalitas residual X2
shapiro.test(resid(M2))
##
## Shapiro-Wilk normality test
##
## data: resid(M2)
## W = 0.93927, p-value = 0.5742
#uji non-heteroskedastisitas X2
bptest(M2,varformula=~fitted.values(M2), studentize=F)
##
## Breusch-Pagan test
##
## data: M2
## BP = 0.061595, df = 1, p-value = 0.804
#uji autokorelasi X2
dwtest(M2, alternative='two.sided')
##
## Durbin-Watson test
##
## data: M2
## DW = 0.50193, p-value = 0.005426
## alternative hypothesis: true autocorrelation is not 0
#uji autokorelasi X2
bgtest(M2)
##
## Breusch-Godfrey test for serial correlation of order up to 1
##
## data: M2
## LM test = 5.4817, df = 1, p-value = 0.01922
#uji glejser
u2=resid(M2)
u2abs=abs(u2)
M2_abs=lm(formula=u2abs~X2, data=data)
summary(M2_abs)
##
## Call:
## lm(formula = u2abs ~ X2, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.3452 -1.7837 -0.6405 1.2837 4.7895
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.4690 3.1460 0.467 0.655
## X2 0.4406 0.8321 0.529 0.613
##
## Residual standard error: 2.631 on 7 degrees of freedom
## Multiple R-squared: 0.03851, Adjusted R-squared: -0.09885
## F-statistic: 0.2804 on 1 and 7 DF, p-value: 0.6128
#Model 3
M3=lm(formula=Y~X3, data=data)
summary(M3)
##
## Call:
## lm(formula = Y ~ X3, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.693 -3.315 -2.592 3.937 9.297
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 27.1873 18.9650 1.434 0.195
## X3 1.1399 0.6369 1.790 0.117
##
## Residual standard error: 5.611 on 7 degrees of freedom
## Multiple R-squared: 0.314, Adjusted R-squared: 0.216
## F-statistic: 3.204 on 1 and 7 DF, p-value: 0.1166
#Pengujian Asumsi
#uji normalitas residual X3
shapiro.test(resid(M3))
##
## Shapiro-Wilk normality test
##
## data: resid(M3)
## W = 0.81761, p-value = 0.03238
#uji non-heteroskedastisitas X3
bptest(M3,varformula=~fitted.values(M3), studentize=F)
##
## Breusch-Pagan test
##
## data: M3
## BP = 0.32297, df = 1, p-value = 0.5698
#uji autokorelasi X3
dwtest(M3, alternative='two.sided')
##
## Durbin-Watson test
##
## data: M3
## DW = 1.117, p-value = 0.1651
## alternative hypothesis: true autocorrelation is not 0
#uji autokorelasi X3
bgtest(M3)
##
## Breusch-Godfrey test for serial correlation of order up to 1
##
## data: M3
## LM test = 0.80317, df = 1, p-value = 0.3701
#uji glejser
u3=resid(M3)
u3abs=abs(u3)
M3_abs=lm(formula=u3abs~X3, data=data)
summary(M3_abs)
##
## Call:
## lm(formula = u3abs ~ X3, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.6039 -1.0955 -1.0751 -0.4915 4.5679
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 13.4360 7.5924 1.77 0.120
## X3 -0.3034 0.2550 -1.19 0.273
##
## Residual standard error: 2.246 on 7 degrees of freedom
## Multiple R-squared: 0.1682, Adjusted R-squared: 0.0494
## F-statistic: 1.416 on 1 and 7 DF, p-value: 0.2729
#Model 4
M4=lm(formula=Y~X1+X2+X3, data=data)
summary(M4)
##
## Call:
## lm(formula = Y ~ X1 + X2 + X3, data = data)
##
## Residuals:
## 1 2 3 4 5 6 7 8 9
## -0.2981 -0.2142 0.4714 -0.4746 -0.9524 0.4761 0.9274 0.7466 -0.6823
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 21.87353 4.07389 5.369 0.00302 **
## X1 0.41277 0.03067 13.460 4.05e-05 ***
## X2 2.20267 0.47199 4.667 0.00550 **
## X3 -0.07895 0.15551 -0.508 0.63330
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8484 on 5 degrees of freedom
## Multiple R-squared: 0.9888, Adjusted R-squared: 0.9821
## F-statistic: 147.1 on 3 and 5 DF, p-value: 2.696e-05
#Pengujian Asumsi
#uji normalitas residual
shapiro.test(resid(M4))
##
## Shapiro-Wilk normality test
##
## data: resid(M4)
## W = 0.93208, p-value = 0.5014
#uji multikollinearitas
vif(M4)
## X1 X2 X3
## 1.403219 3.093549 2.607991
#uji non-heteroskedastisitas
bptest(M4,varformula=~fitted.values(M4), studentize=F)
##
## Breusch-Pagan test
##
## data: M4
## BP = 0.031878, df = 1, p-value = 0.8583
#uji autokorelasi
dwtest(M4, alternative='two.sided')
##
## Durbin-Watson test
##
## data: M4
## DW = 1.6446, p-value = 0.233
## alternative hypothesis: true autocorrelation is not 0
#uji autokorelasi
bgtest(M4)
##
## Breusch-Godfrey test for serial correlation of order up to 1
##
## data: M4
## LM test = 0.51249, df = 1, p-value = 0.4741
#uji glejser
u4=resid(M4)
u4abs=abs(u4)
M4_abs=lm(formula=u4abs~X1+X2+X3, data=data)
summary(M4_abs)
##
## Call:
## lm(formula = u4abs ~ X1 + X2 + X3, data = data)
##
## Residuals:
## 1 2 3 4 5 6 7 8
## -0.24968 -0.37078 -0.06495 -0.01987 0.35744 -0.17650 0.37477 0.12608
## 9
## 0.02347
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.000238 1.548904 0.646 0.547
## X1 0.002391 0.011660 0.205 0.846
## X2 0.035130 0.179452 0.196 0.853
## X3 -0.024936 0.059127 -0.422 0.691
##
## Residual standard error: 0.3226 on 5 degrees of freedom
## Multiple R-squared: 0.04467, Adjusted R-squared: -0.5285
## F-statistic: 0.07793 on 3 and 5 DF, p-value: 0.9692