Getting in data, we have
chemical1 <- c(73, 68, 74, 71, 67)
chemical2 <- c(73, 67, 75, 72, 70)
chemical3 <- c(75, 68, 78, 73, 68)
chemical4 <- c(73, 71, 75, 75, 69)
dat<- data.frame(chemical1,chemical2,chemical3,chemical4)
dat <- stack(dat)
colnames(dat) <- c('Number','Type')
library(GAD)
dat$Type <- as.fixed(dat$Type)
dat$Bolt <- c(rep(seq(1,5),4))
dat$Bolt <- as.fixed(dat$Bolt)
Because we have fixed effect our hypothesis can be written as
The Linear Effects model can be written has;
yij=μ+τi+βj+ϵij
μ is the grand mean value
τi is the fixed effects for treatment i
Where βj is the block effect for j
Where ϵij is the random error
A Randomized Complete Block Design test is being done, where bolt type being blocked (because we are considering bolt type to be a significant source of nuisance variability), and testing chemical types.
model1<- lm(Number~Type+Bolt,dat)
gad(model1)
## Analysis of Variance Table
##
## Response: Number
## Df Sum Sq Mean Sq F value Pr(>F)
## Type 3 12.95 4.317 2.3761 0.1211
## Bolt 4 157.00 39.250 21.6055 2.059e-05 ***
## Residual 12 21.80 1.817
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
The p-value for our model is 0.1211, which is greater than our reference significance level (0.05). So therefore we fail to reject Ho, and thus we conclude that different chemical agents do not have a significant effect on the mean strength
Getting data
AND
Estimating Tau
chem1<- c(73,68,74,71,67)
chem2<- c(73,67,75,72,70)
chem3<- c(75,68,78,73,68)
chem4<- c(73,71,75,75,69)
chems<- c(chem1,chem2,chem3,chem4)
tau1 <- mean(chem1)-mean(chems)
tau2 <- mean(chem2)-mean(chems)
tau3 <- mean(chem3)-mean(chems)
tau4 <- mean(chem4)-mean(chems)
Getting data to get beta_j
Chem5 <- c(73,73,75,73)
Chem6 <- c(68,67,68,71)
Chem7 <- c(74,75,78,75)
Chem8 <- c(71,72,73,75)
Chem9 <- c(67,70,68,69)
Chem10<- c(Chem5,Chem6,Chem7,Chem8,Chem9)
Estimating beta_j, we have
beta1<- mean(Chem5)-mean(Chem10)
beta2<- mean(Chem6)-mean(Chem10)
beta3<- mean(Chem7)-mean(Chem10)
beta4<- mean(Chem8)-mean(Chem10)
beta5<- mean(Chem9)-mean(Chem10)
τ1= -1.15
τ2= -0.35,
τ3= 0.65,
τ4= 0.85,
β1= 1.75,
β2= -3.25,
β3=3.75
β4=1,
β5=-3.25
Getting in data
library(GAD)
batchs<- c(rep(1,5),rep(2,5),rep(3,5),rep(4,5),rep(5,5))
Days<- c(rep(seq(1,5),5))
ingred<- c("A","B","D","C","E","C","E","A","D","B","B","A","C","E","D","D","C","E","B","A","E","D","B","A","C")
obs<- c(8,7,1,7,3,11,2,7,3,8,4,9,10,1,5,6,8,6,6,10,4,2,3,8,8)
dat<- cbind(batchs,Days,ingred,obs)
dat<- as.data.frame(dat)
dat$batchs<- as.fixed(dat$batchs)
dat$Days<- as.fixed(dat$Days)
dat$ingred <- as.fixed(dat$ingred)
Hypothesis:
Null Hypothesis:
Ho:μ1=μ2=μ3=μi
Alternative Hypothesis:
Ha : atleast one μi differs
Because it has fixed effects hence we can write our hypothesis as,
Null Hypothesis:
Ho:τi=0 for all i
Alternative Hypothesis:
Ha : τi≠0 for some i
Writing the Model Equation
Xij=μ+τi+βj+αk+ϵijk
Where,
Xij=Observation
μ=Grand Mean
τi=Effect for "i"
βj= First Block Factor
αk= Second Block Factor
ϵijk= Random Error
model1<- lm(dat$obs~dat$batchs+dat$Days+dat$ingred,data = dat)
anova(model1)
## Analysis of Variance Table
##
## Response: dat$obs
## Df Sum Sq Mean Sq F value Pr(>F)
## dat$batchs 4 15.44 3.860 1.2345 0.3476182
## dat$Days 4 12.24 3.060 0.9787 0.4550143
## dat$ingred 4 141.44 35.360 11.3092 0.0004877 ***
## Residuals 12 37.52 3.127
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Since our p value for our calculated model analysis(0.0004877) for ingredients is lesser than our reference p-value (0.05). we are reject the null hypothesis and stating to conclude that there is a significant effect of the types of ingredients on the mean reaction times of chemical processes.