linear.tools (version 1.3.0)

Manipulate Formulas and Evaluate Marginal Effects

Description

Provides tools to manipulate formulas, such as getting x, y or contrasts from the model/formula, and functions to evaluate and check the marginal effects of a linear model.

Copy Link

Version

Install

install.packages('linear.tools')

Monthly Downloads

Version

1.3.0

License

GPL-3

Maintainer

Fan Yang

Last Published

July 6th, 2016

Functions in linear.tools (1.3.0)

check_names_delete

check whether list names (tobechecked) is within specified sets ('checking')

check_single_numeric

check whether an object is a single numeric number

check_vec_meaningful

check x is a meaningful vector

deleting_wrongeffect

check monotonicity of marginal impacts and re-estimate the model.

effect

evaluate the marginal effects of the selected raw variable on the dependent.

expand_grid

a wrap up of expand.grid() that takes list

focusing_var_coeff

focusing on selected variables in the model, and eliminating impacts from other variables.

get_contrast

get contrast of categorical variables in a model

get_data_from_lm

get raw data from lm or glm

get_model_pair

get a list of model vars with their corresponding coeff vars or raw vars.

get_model_with_coeff

get a list of model variables with their corresponding coeff vars.

get_model_with_raw

get a list of model vars with their corresponding raw vars.

get_valid_rows

identify missing rows for model/formula.

get_x

get x (left hand of var) from model or formula

get_x_all

a unique combinations of model vars, coeff vars and raw vars

get_x_hidden

the underlying function of get_x

get_y

get y (right hand of var)

is_decrease

check whether the vector is decreasing

is_increase

check whether the vector is increasing

Mode

get mode number from a numeric vector

paste_formula

paste a formula as string

sanity_check

check x with various characters

stepwise2

same as step() in R, but able to check marginal effects.

stripGlmLR

make the lm or glm thin

Enter_to_Continue

Enter_to_Continue: wait your response to continue