Scribd
Upload
58 views

Lab 5 EA

The document discusses analyzing normal and chi-squared distributions in R and performing various statistical tests and visualizations on sample data. It generates random normal and chi-squared distributions, performs histograms and QQ plots to compare the sample distributions to theoretical normal distributions, and attempts several statistical tests for normality including the Lillie test. It also explores issues of multicollinearity in sample data by calculating the correlation matrix and checking for high correlations. Linear regression is performed to explore relationships between variables.

Uploaded by

Andrew Trejo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
58 views

Lab 5 EA

The document discusses analyzing normal and chi-squared distributions in R and performing various statistical tests and visualizations on sample data. It generates random normal and chi-squared distributions, performs histograms and QQ plots to compare the sample distributions to theoretical normal distributions, and attempts several statistical tests for normality including the Lillie test. It also explores issues of multicollinearity in sample data by calculating the correlation matrix and checking for high correlations. Linear regression is performed to explore relationships between variables.

Uploaded by

Andrew Trejo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 4

Andrew E Trejo

1226118
Lab 5
NORMALIDAD
>
> #Distribucion Normal
> set.seed(100)
> N <- rnorm(100)
>
> # Distribucion Chi-cuadrado
> set.seed(100)
> C <- rchisq(100, df=5)
>
> # Visualizacion
> par(mfrow=c(1,2))
> hist(N, col="navyblue")
> hist(C, col="navyblue")
>
> # Visualizacion ggplot2
> hoja1 <- data.frame(Normal=N,Chi2=C)
>
> dev.new()
> gN <- ggplot(hoja1, aes(x=Normal))
> gN + geom_histogram(color="white", fill="navyblue", bins = 10)
>
> dev.new()
> gC <- ggplot(hoja1, aes(x=Chi2))
> gC + geom_histogram(color="black", fill="red", bins = 10)
> datos.n <- read.csv(file.choose(),head=T);str(datos.n)
Warning messages:
1: In if (!header) rlabp <- FALSE :
the condition has length > 1 and only the first element will be used
2: In if (header) { :
the condition has length > 1 and only the first element will be used
'data.frame': 474 obs. of 9 variables:
$ fechnac : Factor w/ 462 levels " ","01/02/1951",..: 58 207 286 153 68 329 165
181 45 78 ...
$ genero : int 1 1 2 2 1 1 1 2 2 2 ...
$ educ : int 15 16 12 8 15 15 15 12 15 12 ...
$ catlab : int 3 1 1 1 1 1 1 1 1 1 ...
$ salario : int 57000 40200 21450 21900 45000 32100 36000 21900 27900 24000 ...
$ salini : int 27000 18750 12000 13200 21000 13500 18750 9750 12750 13500 ...
$ tiempemp: int 98 98 98 98 98 98 98 98 98 98 ...
$ expprev : int 144 36 381 190 138 67 114 0 115 244 ...
$ minoria : int 1 1 1 1 1 1 1 1 1 1 ...
> # Visualizacion
> par(mfrow=c(2,2))
> hist(N, col="navyblue")
> hist(datos.n$expprev, col="navyblue")
> qqnorm(N, pch=16)
> qqline(N, col="red")
> qqnorm(datos.n$expprev, pch=16)
> qqline(datos.n$expprev, col="red")
> # Visualizacion ggplot2
> dev.new()
> gN <- ggplot(hoja1, aes(x=Normal))
> gN + geom_histogram(color="white", fill="navyblue", bins = 10)
>
> dev.new()
> gN <- ggplot(datos.n, aes(x=expprev))
> gN + geom_histogram(color="white", fill="navyblue", bins = 10)
>
> dev.new()
> gQQ <- ggplot(hoja1, aes(sample=Normal))
> gQQ + stat_qq() + stat_qq_line(color = "red")
>
> dev.new()
> gQQ <- ggplot(datos.n, aes(sample=expprev))
> gQQ + stat_qq() + stat_qq_line(color = "red")
> dev.new()
Warning messages:
1: In match(x, table, nomatch = 0L) : display list redraw incomplete
2: In unit %in% c("strwidth", "strheight", "strascent", "strdescent") :
display list redraw incomplete
> gQQ <- ggplot(datos.n, aes(sample=expprev))
> gQQ + stat_qq() + stat_qq_line(color = "red")
> dev.new()
> gQQ <- ggplot(datos.n, aes(sample=expprev))
> gQQ + stat_qq()
> install.packages(nortest)
Error in install.packages(nortest) : object 'nortest' not found
> install.packages(nortest)
Error in install.packages(nortest) : object 'nortest' not found
> install.packages("nortest")
--- Please select a CRAN mirror for use in this session ---
Warning: unable to access index for repository
https://cran.revolutionanalytics.com/src/contrib:
cannot open URL 'https://cran.revolutionanalytics.com/src/contrib/PACKAGES'
Warning: unable to access index for repository
https://cran.revolutionanalytics.com/bin/windows/contrib/3.6:
cannot open URL
'https://cran.revolutionanalytics.com/bin/windows/contrib/3.6/PACKAGES'
Warning message:
package �nortest� is not available (for R version 3.6.1)
> library(nortest)
Error in library(nortest) : there is no package called �nortest�
>
> lillie.test(N)
Error in lillie.test(N) : could not find function "lillie.test"
> library(nortest)
Error in library(nortest) : there is no package called �nortest�
> library(nortest)
Error in library(nortest) : there is no package called �nortest�
> # Ho: La distribucion observada se ajusta a la distribucion teorica
>
> library(nortest)
Error in library(nortest) : there is no package called �nortest�
>
> lillie.test(N)
Error in lillie.test(N) : could not find function "lillie.test"
>
> # p-value = 0.1588
>
> # p-value < 0.05, se rechaza Ho, la distribucion no es normal, FALSO
>
> lillie.test(datos.n$expprev)
Error in lillie.test(datos.n$expprev) :
could not find function "lillie.test"
>
> # p-value = 2.2e-16
>
> # p-value < 0.05, se rechaza Ho, la distribucion no es normal, VERDADERO

> install.packages("nortest")
-----------------------------------------------------------
MULTICOLINEALIDAD
$ tiempemp: int 98 98 98 98 98 98 98 98 98 98 ...
$ expprev : int 144 36 381 190 138 67 114 0 115 244 ...
$ minoria : int 1 1 1 1 1 1 1 1 1 1 ...
> cor(datos[,-1])
genero educ catlab salario salini
genero 1.00000000 -0.35598562 -0.377660072 -0.44992300 -0.45667563
educ -0.35598562 1.00000000 0.513853677 0.66055891 0.63319565
catlab -0.37766007 0.51385368 1.000000000 0.78011486 0.75466244
salario -0.44992300 0.66055891 0.780114863 1.00000000 0.88011747
salini -0.45667563 0.63319565 0.754662438 0.88011747 1.00000000
tiempemp -0.06646673 0.04737878 0.005328829 0.08409227 -0.01975347
expprev -0.16485670 -0.25235252 0.062644949 -0.09746693 0.04513563
minoria -0.07566758 -0.13288857 -0.143781245 -0.17733731 -0.15759773
tiempemp expprev minoria
genero -0.066466734 -0.164856699 -0.07566758
educ 0.047378777 -0.252352521 -0.13288857
catlab 0.005328829 0.062644949 -0.14378124
salario 0.084092267 -0.097466926 -0.17733731
salini -0.019753475 0.045135627 -0.15759773
tiempemp 1.000000000 0.002978134 0.04950064
expprev 0.002978134 1.000000000 0.14474651
minoria 0.049500639 0.144746512 1.00000000
> # Redondeamos el resultado
> datos.round <- round(cor(datos[,-1]),2);datos.round
genero educ catlab salario salini tiempemp expprev minoria
genero 1.00 -0.36 -0.38 -0.45 -0.46 -0.07 -0.16 -0.08
educ -0.36 1.00 0.51 0.66 0.63 0.05 -0.25 -0.13
catlab -0.38 0.51 1.00 0.78 0.75 0.01 0.06 -0.14
salario -0.45 0.66 0.78 1.00 0.88 0.08 -0.10 -0.18
salini -0.46 0.63 0.75 0.88 1.00 -0.02 0.05 -0.16
tiempemp -0.07 0.05 0.01 0.08 -0.02 1.00 0.00 0.05
expprev -0.16 -0.25 0.06 -0.10 0.05 0.00 1.00 0.14
minoria -0.08 -0.13 -0.14 -0.18 -0.16 0.05 0.14 1.00
> ------------------------------------------
LINEALIDAD
> datos <- read.csv(file.choose(),head=T);str(datos)
Warning messages:
1: In if (!header) rlabp <- FALSE :
the condition has length > 1 and only the first element will be used
2: In if (header) { :
the condition has length > 1 and only the first element will be used
'data.frame': 474 obs. of 9 variables:
$ fechnac : Factor w/ 462 levels " ","01/02/1951",..: 58 207 286 153 68 329 165
181 45 78 ...
$ genero : int 1 1 2 2 1 1 1 2 2 2 ...
$ educ : int 15 16 12 8 15 15 15 12 15 12 ...
$ catlab : int 3 1 1 1 1 1 1 1 1 1 ...
$ salario : int 57000 40200 21450 21900 45000 32100 36000 21900 27900 24000 ...
$ salini : int 27000 18750 12000 13200 21000 13500 18750 9750 12750 13500 ...
$ tiempemp: int 98 98 98 98 98 98 98 98 98 98 ...
$ expprev : int 144 36 381 190 138 67 114 0 115 244 ...
$ minoria : int 1 1 1 1 1 1 1 1 1 1 ...
>
> # Gr�ficos de dispersi�n
> plot(datos, pch=16)
> ---------------------------------------------------------------------------
AUTOCORRELACION

> # Wallis
> # h-Durbin
> # Breusch-Godfrey
> # Cochrane-Orcutt
>
> # salario = b0 + b1(educ) + b2(tiempemp)
>
> summary(datos.lm)

Call:
lm(formula = salario ~ educ + tiempemp, data = datos.n)

Residuals:
Min 1Q Median 3Q Max
-22432 -7880 -2785 6036 77787

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -25415.26 5415.86 -4.693 3.54e-06 ***
educ 3895.07 204.49 19.048 < 2e-16 ***
tiempemp 89.81 58.63 1.532 0.126
---
Signif. codes: 0 �***� 0.001 �**� 0.01 �*� 0.05 �.� 0.1 � � 1

Residual standard error: 12820 on 471 degrees of freedom


Multiple R-squared: 0.4391, Adjusted R-squared: 0.4368
F-statistic: 184.4 on 2 and 471 DF, p-value: < 2.2e-16

>
> # Residuos estudentizados
> datos.lm.rs <- rstudent(datos.lm)
>
> plot(datos.lm.rs,pch=16,type="b")

You might also like