Linear model for trend analysis

Source: R/trends.R

Performs a linear model for

linear_model(x, y = NULL, conf_level = 0.95)

Arguments

x

numeric vector
y

optional time variable, converted to numeric. If its not provided it will be assumed that all values in X are sequential, regularly measured, and there are no gaps in the measurements.
conf_level

numeric. Level of confidence to be used to calculate the confidence intervals

Value

a tidy data frame with the model results

Examples

x <- runif(100) * 1:100
linear_model(x)
#> # A tibble: 1 × 11
#>    p_value slope conf_low conf_high conf_level intercept r_squared sigma method 
#>      <dbl> <dbl>    <dbl>     <dbl>      <dbl>     <dbl>     <dbl> <dbl> <chr>  
#> 1 2.09e-14 0.550    0.430     0.670       0.95    -0.242     0.451  17.7 Linear…
#> # … with 2 more variables: n <int>, note <chr>
# If measurements were for example taken daily
linear_model(x, y = Sys.Date() + (1:100))
#> # A tibble: 1 × 11
#>    p_value slope conf_low conf_high conf_level intercept r_squared sigma method 
#>      <dbl> <dbl>    <dbl>     <dbl>      <dbl>     <dbl>     <dbl> <dbl> <chr>  
#> 1 2.09e-14 0.550    0.430     0.670       0.95   -10472.     0.451  17.7 Linear…
#> # … with 2 more variables: n <int>, note <chr>