Multivariable Linear Regression

Fits an OLS model via lm(), optionally performing stepwise selection using Akaike's information criterion.

Usage

reg_lm(
  formula,
  data,
  step = FALSE,
  direction = c("both", "forward", "backward"),
  ...
)

Arguments

formula

A formula, e.g. y ~ x1 + x2.

data

A data frame containing the variables in formula.

step

Logical. Whether to perform stepwise selection (TRUE/FALSE). Default is FALSE.

direction

Character: "both" (default), "forward", or "backward".

...

Additional arguments passed to lm() and step().

Value

A named list:

model

Fitted lm object.

coefficient

Tibble of coefficients, standard errors, t-values, p-values, 95% CIs.

model_info

Tibble of fit statistics (r.squared, adj.r.squared, AIC, BIC).

residuals

Tibble of observed, fitted, residuals.

diagnostics

Tibble of key fit metrics (Breusch-Pagan, Durbin-Watson).

formula

Original formula.

input

The original data frame.

Details

When step = TRUE, the function first fits a null model (response only), then uses step() to search for the best combination of terms by AIC.

Examples

set.seed(42)
x1 = rnorm(100); x2 = rnorm(100)
y = 1 + 2*x1 - 3*x2 + rnorm(100, sd = 2)
res = reg_lm(y ~ x1 + x2, data = data.frame(y, x1, x2))
res$coefficient
#> # A tibble: 3 × 7
#>   term        estimate std.error statistic  p.value conf.low conf.high
#>   <chr>          <dbl>     <dbl>     <dbl>    <dbl>    <dbl>     <dbl>
#> 1 (Intercept)     1.00     0.204      4.92 3.48e- 6    0.599      1.41
#> 2 x1              1.71     0.196      8.75 6.72e-14    1.32       2.10
#> 3 x2             -2.83     0.225    -12.5  4.91e-22   -3.28      -2.38