COMMunication
RESearch.NET

library(tidyverse)
library(lavaan)
cormat = lav_matrix_lower2full(c(1,
                                 0.418613, 1,
                                 0.4153223, 0.2652651, 1,
                                 0.2670484, 0.4315334, 0.2709399, 1))
colnames(cormat) <- rownames(cormat) <- c("X1","X2","X3","Y")
cormat

n_mean = matrix( c(rep(254,4),4.066, 4.024, 4.033, 3.919), 
                 ncol=4, nrow = 2, byrow = T )
n_mean
n_mean <- n_mean %>%
  `colnames<-`(c("X1","X2","X3","Y")) %>% 
  `rownames<-`(c("n", "mean"))
n_mean  



# Install the MASS package if not already installed
# install.packages("MASS")
# Load the package
library(MASS)
library(lavaan)

#01 데이터 입력 ####
#AC:개인이 성공 성취도, 
#IN:가계의 수입, 
#AB:개인의 능력, 
#SU:개인의성공열정
#covMat  #공분산 행렬로 결과 확인 

# 1. Define the mean vector (for 3 variables)
mns <- c(100, 120, 150, 130)
# 2. Define the covariance matrix (must be symmetric and positive-definite)
# Example: 3x3 covariance matrix
cv_mat <- lav_matrix_lower2full(
               c(26.5, 
                 21.5, 37.9, 
                 23.5, 25.2, 41.7,
                 16.5, 14.6, 14.5, 15.8))
colnames(cvMat) <- rownames(cvMat) <- c("AC", "IN", "AB","SU")

# 3. Simulate the data (e.g., 100 observations)
n_observations <- 300
sim_d <- mvrnorm(n = n_observations, mu = mns, Sigma = cv_mat)
sim_data <- round(sim_d,0)
# View the first few rows of the simulated data
colnames(sim_data) <- c("AC", "IN", "AB","SU")
head(sim_data)
str(sim_data)

PathModel <- '
SU ~ a*AB +b*IN
AC ~ c*SU + d*AB
AB ~~ IN
'
#03 분석진행은 sem을 이용하여 한다. ####
# sem(모델, sample.cov= 공분산 행렬, sample.nobs=샘플의 개수)
#fitM <- sem(PathModel, sample.cov = covMat, sample.nobs = 300,
#            fixed.x = FALSE)
fitM <- sem(PathModel, data=sim_data)
#04 결과보기 ####
summary(fitM, standardized=TRUE,fit.measures=TRUE, rsquare=T)

# install.packages("semPlot")
library(semPlot)
semPaths(fitM, whatLabels = "std", curveAdjacent = TRUE, 
         style = "lisrel",
         intAtSide=TRUE, intercepts = TRUE, rotation = 2, 
         layout = "tree3",
         edge.label.cex = 1, 
         edge.label.position=1, edge.color = "black",
         color=list(lat="tomato", man="grey"),
         curvature=3,
         exoVar = TRUE,exoCov = TRUE)

data(PoliticalDemocracy)
PoliticalDemocracy
sem.model <- "
  # measurement model
    ind60 =~ x1 + x2 + x3
    dem60 =~ y1 + y2 + y3 + y4
    dem65 =~ y5 + y6 + y7 + y8
  # regressions
    dem60 ~ ind60
    dem65 ~ ind60 + dem60
  # residual correlations
    y1 ~~ y5
    y2 ~~ y4 + y6
    y3 ~~ y7
    y4 ~~ y8
    y6 ~~ y8
"

# Estimate model 
fit1 <- sem(model = sem.model, data = PoliticalDemocracy)

# show summary result 
summary(fit1, fit.measures=TRUE, standardized=TRUE)

library(lavaan)
library(semPlot)
library(OpenMx)
library(tidyverse)
library(knitr)
library(kableExtra)
library(GGally)

model <-'
mpg ~ hp + gear + cyl + disp + carb + am + wt
hp ~ cyl + disp + carb
'
fit.mod <- sem(model, data=mtcars)
summary(fit.mod, standardize=T, rsquare=T)


dat <- read.csv("http://commres.net/_media/r/eating.disorder.sim.csv")
head(dat)
n <- nrow(dat)
S <- cov(dat)
cor <- cor(dat)

# 
# xSelf-Efficacy = BMI + Self-Esteem + XDisturbance
# Bulimic Symptoms = BMI + Self-Esteem + XSelf-Efficacy + XDisturbance
# Restrictive Symptoms = BMI + Self-Esteem + xSelf-Efficacy + xDisturbance
# Overall Risk = BMI + Self-Esteem + xSelf-Efficacy + Acculturation + xDisturbance
#  

ex1 <- "           #opening a quote
  # Tilda ~ : Regression 
  # M ~ X regression (X predicts M)
  # Each line corresponds to an equation 
  # Disturbance is automatically included for each regression 
  # (i.e. no extra term needed)
  
  DietSE ~ BMI + SelfEsteem      #DietSE is predicted by BMI and SelfEsteem 
  Bulimia ~ DietSE + BMI + SelfEsteem
  Restrictive ~ DietSE + BMI + SelfEsteem
  Risk ~ DietSE + BMI + SelfEsteem + Accu
"  
ex1.fit <- sem(ex1, data=dat)
summary(ex1.fit, standardize=T, rsquare=T)
ex1b.fit <- sem(model = ex1, sample.cov = S, sample.nobs = n)
summary(ex1b.fit)
library(tidyverse)
library(lavaan)
cormat = lav_matrix_lower2full(c(1,
                                 0.418613, 1,
                                 0.4153223, 0.2652651, 1,
                                 0.2670484, 0.4315334, 0.2709399, 1))
colnames(cormat) <- rownames(cormat) <- c("X1","X2","X3","Y")
cormat

n_mean = matrix( c(rep(254,4),4.066, 4.024, 4.033, 3.919), 
                 ncol=4, nrow = 2, byrow = T )
n_mean
n_mean <- n_mean %>%
  `colnames<-`(c("X1","X2","X3","Y")) %>% 
  `rownames<-`(c("n", "mean"))
n_mean  



# Install the MASS package if not already installed
# install.packages("MASS")
# Load the package
library(MASS)
library(lavaan)

#01 데이터 입력 ####
#AC:개인이 성공 성취도, 
#IN:가계의 수입, 
#AB:개인의 능력, 
#SU:개인의성공열정
#covMat  #공분산 행렬로 결과 확인 

# 1. Define the mean vector (for 3 variables)
mns <- c(100, 120, 150, 130)
# 2. Define the covariance matrix (must be symmetric and positive-definite)
# Example: 3x3 covariance matrix
cv_mat <- lav_matrix_lower2full(
               c(26.5, 
                 21.5, 37.9, 
                 23.5, 25.2, 41.7,
                 16.5, 14.6, 14.5, 15.8))
colnames(cvMat) <- rownames(cvMat) <- c("AC", "IN", "AB","SU")

# 3. Simulate the data (e.g., 100 observations)
n_observations <- 300
sim_d <- mvrnorm(n = n_observations, mu = mns, Sigma = cv_mat)
sim_data <- round(sim_d,0)
# View the first few rows of the simulated data
colnames(sim_data) <- c("AC", "IN", "AB","SU")
head(sim_data)
str(sim_data)

PathModel <- '
SU ~ a*AB +b*IN
AC ~ c*SU + d*AB
AB ~~ IN
'
#03 분석진행은 sem을 이용하여 한다. ####
# sem(모델, sample.cov= 공분산 행렬, sample.nobs=샘플의 개수)
#fitM <- sem(PathModel, sample.cov = covMat, sample.nobs = 300,
#            fixed.x = FALSE)
fitM <- sem(PathModel, data=sim_data)
#04 결과보기 ####
summary(fitM, standardized=TRUE,fit.measures=TRUE, rsquare=T)

# install.packages("semPlot")
library(semPlot)
semPaths(fitM, whatLabels = "std", curveAdjacent = TRUE, 
         style = "lisrel",
         intAtSide=TRUE, intercepts = TRUE, rotation = 2, 
         layout = "tree3",
         edge.label.cex = 1, 
         edge.label.position=1, edge.color = "black",
         color=list(lat="tomato", man="grey"),
         curvature=3,
         exoVar = TRUE,exoCov = TRUE)

data(PoliticalDemocracy)
PoliticalDemocracy
sem.model <- "
  # measurement model
    ind60 =~ x1 + x2 + x3
    dem60 =~ y1 + y2 + y3 + y4
    dem65 =~ y5 + y6 + y7 + y8
  # regressions
    dem60 ~ ind60
    dem65 ~ ind60 + dem60
  # residual correlations
    y1 ~~ y5
    y2 ~~ y4 + y6
    y3 ~~ y7
    y4 ~~ y8
    y6 ~~ y8
"

# Estimate model 
fit1 <- sem(model = sem.model, data = PoliticalDemocracy)

# show summary result 
summary(fit1, fit.measures=TRUE, standardized=TRUE)

library(lavaan)
library(semPlot)
library(OpenMx)
library(tidyverse)
library(knitr)
library(kableExtra)
library(GGally)

model <-'
mpg ~ hp + gear + cyl + disp + carb + am + wt
hp ~ cyl + disp + carb
'
fit.mod <- sem(model, data=mtcars)
summary(fit.mod, standardize=T, rsquare=T)


dat <- read.csv("http://commres.net/_media/r/eating.disorder.sim.csv")
head(dat)
n <- nrow(dat)
S <- cov(dat)
cor <- cor(dat)

# 
# xSelf-Efficacy = BMI + Self-Esteem + XDisturbance
# Bulimic Symptoms = BMI + Self-Esteem + XSelf-Efficacy + XDisturbance
# Restrictive Symptoms = BMI + Self-Esteem + xSelf-Efficacy + xDisturbance
# Overall Risk = BMI + Self-Esteem + xSelf-Efficacy + Acculturation + xDisturbance
#  

ex1 <- "           #opening a quote
  # Tilda ~ : Regression 
  # M ~ X regression (X predicts M)
  # Each line corresponds to an equation 
  # Disturbance is automatically included for each regression 
  # (i.e. no extra term needed)
  
  DietSE ~ BMI + SelfEsteem      #DietSE is predicted by BMI and SelfEsteem 
  Bulimia ~ DietSE + BMI + SelfEsteem
  Restrictive ~ DietSE + BMI + SelfEsteem
  Risk ~ DietSE + BMI + SelfEsteem + Accu
"  
ex1.fit <- sem(ex1, data=dat)
summary(ex1.fit, standardize=T, rsquare=T)
ex1b.fit <- sem(model = ex1, sample.cov = S, sample.nobs = n)
summary(ex1b.fit)

> rm(list =ls())
> library(readr)
> df <- read.csv("http://commres.net/_media/r/plannedbehavior.csv")
> colnames(df) <- c("att", "nor", "cont", "int", "beh")
> head(df)
   att  nor cont  int  beh
1 2.31 2.31 2.03 2.50 2.62
2 4.66 4.01 3.63 3.99 3.64
3 3.85 3.56 4.20 4.35 3.83
4 4.24 2.25 2.84 1.51 2.25
5 2.91 3.31 2.40 1.45 2.00
6 2.99 2.51 2.95 2.59 2.20
> 
> attach(df)
>
> lm.ba.01 <- lm(beh ~ att + int)
> lm.ba.02 <- lm(beh ~ int)
> lm.ba.03 <- lm(int ~ att)
> lm.ba.04 <- lm(att ~ int)
> lm.ba.05 <- lm(beh ~ att)
> 
> abc <- summary(lm.ba.01)$r.square
> ab <-  summary(lm.ba.02)$r.square
> bc <-  summary(lm.ba.05)$r.square
> ab+bc
[1] 0.2602023
> abc
[1] 0.1989125
> ab+bc - abc
[1] 0.0612898
> 
> int.reg <- lm.ba.03$fitted.values - mean(int)
> # be/abef 부분
> int.res <- summary(lm.ba.03)$residuals
> # af/abef 부분
> 
> lm.ba.031 <- lm(beh~int.reg)
> lm.ba.032 <- lm(beh~int.res)
> b2 <- summary(lm.ba.031)$r.square
> a2 <- summary(lm.ba.032)$r.square
> a2
[1] 0.1369399
> b2
[1] 0.06197255
> a2+b2
[1] 0.1989125
> ab
[1] 0.1982297
> 

graph TD
  A(**mermaid**)-->B((__plugin__))
  A-->C(((//for//)))
  B-->D[["[[https://www.dokuwiki.org/dokuwiki|Dokuwiki]]"]]
  C-->D

\begin{eqnarray*} & & P(A \mid B) = \dfrac{P(A \cap B)}{P(B)}\\ & & P(B \mid A) = \dfrac{P(B \cap A)}{P(A)}\\ \\ & & P(B \vert A) \;\; \text{ vs. } \;\; P(B \mid A) \\ & & P(A \cap B) = P(A \mid B) * P(B) \\ & & P(B \cap A) = P(B \mid A) * P(A) \\ & & P(A \cap B) = P(A, B) \\ \\ & & \frac{3}{4 \pi} \sqrt{4 \cdot x^2 12} \\ & & \lim_{n \to \infty} \sum_{k=1}^n \frac{1}{k^2} = \frac{\pi^2}{6} \\ & & {\it f}(x) = \frac{1}{\sqrt{x} x^2} \\ & & e^{i \pi} + 1 = 0\; \end{eqnarray*}

\begin{align*} & \;\;\;\; \sum{(Y_i - \hat{Y_i})^2} \\ &= \sum{(Y_i - (a + bX_i))^2} \;\;\; \because \hat{Y_i} = a + bX_i \\ &= \text{SSE or SS.residual} \;\;\; \text{(and this should be the least value.)} \\ \end{align*}