CODEWITHSUNDEEP
rggplot2plotvisualization
clause_yl
clause_yl

Reputation: 21

How to make my plot smoother in R using plot function?

This is my code and my final graph result. But I don't know how to make the line smoother. Are there any suggestions? I tried loess and spline but I don't know how I should apply them on my code.

p2 <- seq(.1587, .18144, .00756)
np2 <- length(p2)
power <- seq(.4, .9, .1)
npower <- length(power)
samsize <- array(numeric(np2*npower), dim = c(np2,npower))
  for (i in 1:npower){
    for (j in 1:np2){
      result <- power.prop.test(n = NULL, p1 =.1511, p2 = p2[j], 
                                sig.level = .05, power = power[i],
                                alternative = c("two.sided","one.sided"),
                                strict = FALSE, tol = .Machine$double.eps^.25)
        samsize[j,i] <- ceiling(result$n)
      }
    }

    xrange <- range(p2)
    yrange <- round(range(samsize))
    colors <- rainbow(length(power))
    plot(xrange, yrange, type = "n",
         xlab = "Target Percentage (p2)",
         ylab = "Sample Size (n)")

    for (i in 1:npower){
      lines(p2, samsize[,i], lwd = 2, col=colors[i])
    }

    abline(h=seq(0,50000,5000), v=seq(xrange[1],xrange[2], .002),lty=20,col="gray89")
    title("Sample Size Estimation for Proportion Test")
    legend("topright", title= "Power", as.character(power),
           fill=colors)

Thank you!

enter image description here

Upvotes: 1

Views: 128

Answers (2)

at80
at80

Reputation: 780

You just need to convert your array to a dataframe that ggplot can handle. Then you can use geom_smooth.

df <- data.frame(samsize)
colnames(df) <- as.character(power)

df <- tidyr::gather(df)
df$p2 <- p2

ggplot() + 
  geom_smooth(data=df, aes(x=p2, y=value, color=key)) +
  scale_color_discrete(name = 'Power') +
  labs(x = 'Target Percentage (p2)', y = 'Sample Size (n)', title="Sample Size Estimation for Proportion Test") +
  theme_bw() +
  theme(plot.title = element_text(hjust = 0.5))

Upvotes: 1

Maurits Evers
Maurits Evers

Reputation: 50668

You could increase the number of points for p2 at which you calculate the sample size.

For example, if you set

p2 <- seq(.1587, .18144, length.out = 100)

then you get

enter image description here

Upvotes: 2

Related Questions