34  Regression models: interactions using indexing

polite <- read_csv("data/winter2012/polite.csv")

f0_bm_int2 <- brm(
  f0mn ~ 0 + gender:attitude,
  family = gaussian,
  data = polite,
  cores = 4,
  seed = 7123,
  file = "cache/ch-regression-interaction-index_f0_bm_int2"
)

summary(f0_bm_int2)

 Family: gaussian 
  Links: mu = identity; sigma = identity 
Formula: f0mn ~ 0 + gender:attitude 
   Data: polite (Number of observations: 212) 
  Draws: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
         total post-warmup draws = 4000

Regression Coefficients:
                    Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
genderF:attitudeinf   256.61      5.16   246.73   266.76 1.00     5037     2982
genderM:attitudeinf   137.23      5.87   125.37   148.66 1.00     5491     2936
genderF:attitudepol   238.93      5.08   229.06   248.85 1.00     5191     2912
genderM:attitudepol   126.28      5.82   114.88   137.71 1.00     5455     3018

Further Distributional Parameters:
      Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sigma    39.13      1.94    35.51    43.18 1.00     4975     3159

Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
and Tail_ESS are effective sample size measures, and Rhat is the potential
scale reduction factor on split chains (at convergence, Rhat = 1).

conditional_effects(f0_bm_int2, "gender:attitude")

f0_bm_int2_draws <- as_draws_df(f0_bm_int2)

f0_bm_int2_draws <- f0_bm_int2_draws |> 
  mutate(
    f_pol_inf = `b_genderF:attitudepol` - `b_genderF:attitudeinf`,
    m_pol_inf = `b_genderM:attitudepol` - `b_genderM:attitudeinf`,
  )

library(posterior)

This is posterior version 1.6.1

Attaching package: 'posterior'

The following objects are masked from 'package:stats':

    mad, sd, var

The following objects are masked from 'package:base':

    %in%, match

f0_bm_int2_draws |>
  summarise(
    mean_diff = mean(m_pol_inf), sd_diff = sd(m_pol_inf),
    lo_diff = quantile2(m_pol_inf, probs = 0.025), hi_diff = quantile2(m_pol_inf, probs = 0.975)
  ) |>
  round()