- main parameters: a, v_perc_acc, t, z, v_Intercept
- data within each study are treated as having come from single participant
- plot last figure here [Slab + interval stats and geoms • ggdist](https://mjskay.github.io/ggdist/articles/slabinterval.html)
![[hddm_aggregate.png|1400]]
# boundary
```r
b = 0.04, 95% HPD [−0.45, 0.52]
m1 <- brm(a | mi(a_sd) ~ condition + (1 + condition | Study), data = dt3, iter = 10000)
# effect of condition
b = 0.05 [-0.26, 0.33]
> m1
Family: gaussian
Links: mu = identity; sigma = identity
Formula: a | mi(a_sd) ~ condition + (1 + condition | Study)
Data: dt3 (Number of observations: 12)
Draws: 4 chains, each with iter = 10000; warmup = 5000; thin = 1;
total post-warmup draws = 20000
Group-Level Effects:
~Study (Number of levels: 6)
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sd(Intercept) 0.59 0.30 0.25 1.37 1.03 96 360
sd(condition) 0.18 0.16 0.01 0.59 1.08 3122 4294
cor(Intercept,condition) -0.33 0.52 -0.96 0.87 1.02 208 6542
Population-Level Effects:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
Intercept 5.02 0.28 4.49 5.62 1.02 147 2836
condition 0.05 0.14 -0.26 0.33 1.09 4375 3707
Family Specific Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sigma 0.14 0.12 0.02 0.47 1.15 17 14
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).
```
# weight on accuracy
```r
m2 <- brm(b1 | mi(b1_sd) ~ condition + (1 + condition | Study), data = dt3, iter = 10000)
# effect of condition
b = 0.14 [0.01, 0.29]
> m2
Family: gaussian
Links: mu = identity; sigma = identity
Formula: b1 | mi(b1_sd) ~ condition + (1 + condition | Study)
Data: dt3 (Number of observations: 12)
Draws: 4 chains, each with iter = 10000; warmup = 5000; thin = 1;
total post-warmup draws = 20000
Group-Level Effects:
~Study (Number of levels: 6)
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sd(Intercept) 0.21 0.14 0.03 0.51 1.17 16 1433
sd(condition) 0.08 0.07 0.00 0.26 1.02 596 6807
cor(Intercept,condition) -0.27 0.56 -0.97 0.90 1.05 52 319
Population-Level Effects:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
Intercept 0.20 0.11 0.01 0.42 1.14 19 92
condition 0.14 0.07 0.01 0.29 1.05 1876 5090
Family Specific Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sigma 0.07 0.05 0.02 0.20 1.12 21 8
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).
```
![[hddm_aggregate_brms_posterior.png|1200]]