I'm trying to create a program with R to calculate manually b0 and b1 in Simple Linear Regression with Least Square Method.
regression=function(num,x,y)
switch(num,
b1 = {n = 5
b = (n*sum(x*y)-sum(x)*sum(y))/(n*sum(x^2)-sum(x)^2)
print(b)},
b0 = {n = 5
b = (n*sum(x*y)-sum(x)*sum(y))/(n*sum(x^2)-sum(x)^2)
a = mean(y)-b1*mean(x)
print(a)}
)
x = c(1, 2, 3, 4, 5)
y = c(2, 1, 4, 5, 3)
regression(b1, x, y)
regression(b0, x, y)
But it fails
A simpler way of defining your function is as follows,
regression=function(num,x,y){
n=num
b1 = (n*sum(x*y)-sum(x)*sum(y))/(n*sum(x^2)-sum(x)^2)
b0=mean(y)- b1*mean(x)
return(c(b0,b1))
}
With this, you can get a vector containing your b0 and b1. In the code below, I have shown how you can access this and plot the resulting regression line.
x = c(1, 2, 3, 4, 5)
y = c(2, 1, 4, 5, 3)
b0<-regression(5,x,y)[1]
b1<-regression(5,x,y)[2]
regression_line<-b0+b1*x
plot(x,y)
lines(regression_line)
Two issues.
b0
and b1
don't exist when you call the function, so you can't pass them in as arguments---you can pass them in as strings, which is what switch
expects. So when you call regression
, call it as regression("b1", x, y)
or regression("b0", x, y)
.
In the b0 = {...}
section of code, you call an intermediate result b
, but later try to reference b1
. Again, b1
doesn't exist, so call your intermediate result b1
, not b
.
Address those issues, and I think your function will work just fine :)
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.