LM: Analysis: Linear, quadratic, quadratic inverse, cubic and quartic

Description

Linear, quadratic, quadratic inverse, cubic and quartic regression.

Usage

LM(
  trat,
  resp,
  degree = NA,
  sample.curve = 1000,
  ylab = "Dependent",
  xlab = "Independent",
  error = "SE",
  ic = FALSE,
  fill.ic = "gray70",
  alpha.ic = 0.5,
  point = "all",
  r2 = "all",
  theme = theme_classic(),
  legend.position = "top",
  width.bar = NA,
  scale = "none",
  textsize = 12,
  pointsize = 4.5,
  linesize = 0.8,
  linetype = 1,
  pointshape = 21,
  fillshape = "gray",
  colorline = "black",
  round = NA,
  xname.formula = "x",
  yname.formula = "y",
  comment = NA,
  fontfamily = "sans"
)

Value

The function returns a list containing the coefficients and their respective values of p; statistical parameters such as AIC, BIC, pseudo-R2, RMSE (root mean square error); largest and smallest estimated value and the graph using ggplot2 with the equation automatically.

Arguments

trat: Numeric vector with dependent variable.
resp: Numeric vector with independent variable.
degree: degree of the polynomial (0.5, 1, 2, 3 or 4)
sample.curve: Provide the number of observations to simulate curvature (default is 1000)
ylab: Dependent variable name (Accepts the expression() function)
xlab: Independent variable name (Accepts the expression() function)
error: Error bar (It can be SE - default, SD or FALSE)
ic: Add interval of confidence
fill.ic: Color interval of confidence
alpha.ic: confidence interval transparency level
point: defines whether you want to plot all points ("all") or only the mean ("mean")
r2: coefficient of determination of the mean or all values (default is all)
theme: ggplot2 theme (default is theme_classic())
legend.position: legend position (default is "top")
width.bar: Bar width
scale: Sets x scale (default is none, can be "log")
textsize: Font size
pointsize: shape size
linesize: line size
linetype: line type
pointshape: format point (default is 21)
fillshape: Fill shape
colorline: Color lines
round: round equation
xname.formula: Name of x in the equation
yname.formula: Name of y in the equation
comment: Add text after equation
fontfamily: Font family

Author

Gabriel Danilo Shimizu

Leandro Simoes Azeredo Goncalves

Details

The linear model is defined by: $$y = \beta_0 + \beta_1\cdot x$$ The quadratic model is defined by: $$y = \beta_0 + \beta_1\cdot x + \beta_2\cdot x^2$$ The quadratic inverse model is defined by: $$y = \beta_0 + \beta_1\cdot x + \beta_2\cdot x^{0.5}$$ The cubic model is defined by: $$y = \beta_0 + \beta_1\cdot x + \beta_2\cdot x^2 + \beta_3\cdot x^3$$ The quartic model is defined by: $$y = \beta_0 + \beta_1\cdot x + \beta_2\cdot x^2 + \beta_3\cdot x^3+ \beta_4\cdot x^4$$

Examples

Run this code

library(AgroReg)
data("aristolochia")
attach(aristolochia)
LM(trat,resp, degree = 3)

Run the code above in your browser using DataLab