How do you get the equation of polynomial regression using ggplot and R?

Question

I've been trying to plot some data and add a third order polynomial trend line. I'm using the lm() and coef() functions to obtain a model for my data and then plot with ggplot.

Here is the relevant code:

Bytes is the name of my dependent variable (y). Resolution is the name of my independent variable (x).

model = lm(Bytes~poly(Resolution, 3), averaged_frame) print(coef(model)) library(ggplot2) plot = ggplot(averaged_frame, mapping = aes(x = Resolution, y = Bytes)) + geom_point() + stat_smooth(method="lm", formula = y~poly(x, 3), se = FALSE)

However, when I take the coefficients from print(coef(model)):

 (Intercept) poly(Resolution, 3)1 poly(Resolution, 3)2 poly(Resolution, 3)3 0.3392046 0.3686154 0.2504288 0.1274490

And transpose them into a function:

f(x)=0.1274490x^3+0.2504288x^2+0.3686154x+0.3392046

And graph it, it looks completely different to the plot on ggplot.

I'm wondering what I am doing wrong here or how I might get an equation from the line in ggplot.

You should use raw = TRUE in poly. That's why you get a different curve. — Stéphane Laurent
– Stéphane Laurent, Commented May 13, 2023 at 5:54

Stéphane Laurent · Accepted Answer · 2023-05-13 05:52:36Z

Here is an example with the ggpmisc package. This is for an order one regression, you have to adapt to your case.

df <- data.frame(x = c(1:100)) df$y <- 20 + 30 * df$x + rnorm(100, sd = 80) library(ggpmisc) my_formula <- y ~ x myformat <- "y = %s + %s x --- R²: %s" ggplot(df, aes(x, y)) + geom_point() + geom_smooth(method = "lm", formula = my_formula, se = FALSE) + stat_poly_eq( formula = my_formula, output.type = "numeric", mapping = aes( label = sprintf( myformat, formatC(after_stat(coef.ls)[[1]][[1, "Estimate"]]), formatC(after_stat(coef.ls)[[1]][[2, "Estimate"]]), formatC(stat(r.squared))), ), vstep = 0.1, size = 10 )

Collectives™ on Stack Overflow

How do you get the equation of polynomial regression using ggplot and R?

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related