# Estimating Monotonic Effects with brms

## Introduction

This vignette is about monotonic effects, a special way of handling discrete predictors that are on an ordinal or higher scale (Bürkner & Charpentier, in review). A predictor, which we want to model as monotonic (i.e., having a monotonically increasing or decreasing relationship with the response), must either be integer valued or an ordered factor. As opposed to a continuous predictor, predictor categories (or integers) are not assumed to be equidistant with respect to their effect on the response variable. Instead, the distance between adjacent predictor categories (or integers) is estimated from the data and may vary across categories. This is realized by parameterizing as follows: One parameter, $$b$$, takes care of the direction and size of the effect similar to an ordinary regression parameter. If the monotonic effect is used in a linear model, $$b$$ can be interpreted as the expected average difference between two adjacent categories of the ordinal predictor. An additional parameter vector, $$\zeta$$, estimates the normalized distances between consecutive predictor categories which thus defines the shape of the monotonic effect. For a single monotonic predictor, $$x$$, the linear predictor term of observation $$n$$ looks as follows:

$\eta_n = b D \sum_{i = 1}^{x_n} \zeta_i$

The parameter $$b$$ can take on any real value, while $$\zeta$$ is a simplex, which means that it satisfies $$\zeta_i \in [0,1]$$ and $$\sum_{i = 1}^D \zeta_i = 1$$ with $$D$$ being the number of elements of $$\zeta$$. Equivalently, $$D$$ is the number of categories (or highest integer in the data) minus 1, since we start counting categories from zero to simplify the notation.

## A Simple Monotonic Model

A main application of monotonic effects are ordinal predictors that can be modeled this way without falsely treating them either as continuous or as unordered categorical predictors. In Psychology, for instance, this kind of data is omnipresent in the form of Likert scale items, which are often treated as being continuous for convenience without ever testing this assumption. As an example, suppose we are interested in the relationship of yearly income (in $) and life satisfaction measured on an arbitrary scale from 0 to 100. Usually, people are not asked for the exact income. Instead, they are asked to rank themselves in one of certain classes, say: ‘below 20k’, ‘between 20k and 40k’, ‘between 40k and 100k’ and ‘above 100k’. We use some simulated data for illustration purposes. income_options <- c("below_20", "20_to_40", "40_to_100", "greater_100") income <- factor(sample(income_options, 100, TRUE), levels = income_options, ordered = TRUE) mean_ls <- c(30, 60, 70, 75) ls <- mean_ls[income] + rnorm(100, sd = 7) dat <- data.frame(income, ls) We now proceed with analyzing the data modeling income as a monotonic effect. fit1 <- brm(ls ~ mo(income), data = dat) The summary methods yield summary(fit1)  Family: gaussian Links: mu = identity; sigma = identity Formula: ls ~ mo(income) Data: dat (Number of observations: 100) Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1; total post-warmup samples = 4000 Population-Level Effects: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS Intercept 31.14 1.55 28.07 34.17 1.00 2536 2583 moincome 14.75 0.71 13.36 16.15 1.00 2508 2472 Simplex Parameters: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS moincome1 0.64 0.04 0.56 0.73 1.00 3106 2493 moincome1 0.24 0.04 0.15 0.33 1.00 3605 2749 moincome1 0.12 0.04 0.04 0.19 1.00 3169 2052 Family Specific Parameters: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS sigma 7.08 0.52 6.14 8.17 1.00 3020 2289 Samples were drawn 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). plot(fit1, pars = "simo") plot(conditional_effects(fit1)) The distributions of the simplex parameter of income, as shown in the plot method, demonstrate that the largest difference (about 70% of the difference between minimum and maximum category) is between the first two categories. Now, let’s compare of monotonic model with two common alternative models. (a) Assume income to be continuous: dat$income_num <- as.numeric(dat$income) fit2 <- brm(ls ~ income_num, data = dat) summary(fit2)  Family: gaussian Links: mu = identity; sigma = identity Formula: ls ~ income_num Data: dat (Number of observations: 100) Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1; total post-warmup samples = 4000 Population-Level Effects: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS Intercept 23.16 2.43 18.41 27.83 1.00 4017 3319 income_num 14.54 0.87 12.89 16.26 1.00 4071 3113 Family Specific Parameters: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS sigma 9.14 0.66 7.95 10.53 1.00 3641 2923 Samples were drawn using sampling(NUTS). For each parameter, Bulk_ESS and Tail_ESS are effective sample size measures, and Rhat is the potential scale reduction factor on split chains (at convergence, Rhat = 1). or (b) Assume income to be an unordered factor: contrasts(dat$income) <- contr.treatment(4)
fit3 <- brm(ls ~ income, data = dat)
summary(fit3)
 Family: gaussian
Links: mu = identity; sigma = identity
Formula: ls ~ income
Data: dat (Number of observations: 100)
Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
total post-warmup samples = 4000

Population-Level Effects:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
Intercept    30.84      1.59    27.76    33.89 1.00     2860     2355
income2      28.82      2.25    24.51    33.31 1.00     3213     2670
income3      39.40      2.00    35.46    43.40 1.00     2906     2868
income4      44.69      2.18    40.34    48.93 1.00     3070     2899

Family Specific Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sigma     7.07      0.51     6.15     8.13 1.00     3778     2941

Samples were drawn 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).

We can easily compare the fit of the three models using leave-one-out cross-validation.

loo(fit1, fit2, fit3)
Output of model 'fit1':

Computed from 4000 by 100 log-likelihood matrix

Estimate   SE
elpd_loo   -339.5  7.0
p_loo         4.8  0.8
looic       678.9 14.0
------
Monte Carlo SE of elpd_loo is 0.0.

All Pareto k estimates are good (k < 0.5).
See help('pareto-k-diagnostic') for details.

Output of model 'fit2':

Computed from 4000 by 100 log-likelihood matrix

Estimate   SE
elpd_loo   -364.3  7.1
p_loo         3.1  0.7
looic       728.6 14.3
------
Monte Carlo SE of elpd_loo is 0.0.

All Pareto k estimates are good (k < 0.5).
See help('pareto-k-diagnostic') for details.

Output of model 'fit3':

Computed from 4000 by 100 log-likelihood matrix

Estimate   SE
elpd_loo   -339.5  6.9
p_loo         4.9  0.8
looic       679.0 13.9
------
Monte Carlo SE of elpd_loo is 0.0.

All Pareto k estimates are good (k < 0.5).
See help('pareto-k-diagnostic') for details.

Model comparisons:
elpd_diff se_diff
fit1   0.0       0.0
fit3  -0.1       0.3
fit2 -24.9       5.1  

The monotonic model fits better than the continuous model, which is not surprising given that the relationship between income and ls is non-linear. The monotonic and the unordered factor model have almost identical fit in this example, but this may not be the case for other data sets.

## Setting Prior Distributions

In the previous monotonic model, we have implicitly assumed that all differences between adjacent categories were a-priori the same, or formulated correctly, had the same prior distribution. In the following, we want to show how to change this assumption. The canonical prior distribution of a simplex parameter is the Dirichlet distribution, a multivariate generalization of the beta distribution. It is non-zero for all valid simplexes (i.e., $$\zeta_i \in [0,1]$$ and $$\sum_{i = 1}^D \zeta_i = 1$$) and zero otherwise. The Dirichlet prior has a single parameter $$\alpha$$ of the same length as $$\zeta$$. The higher $$\alpha_i$$ the higher the a-priori probability of higher values of $$\zeta_i$$. Suppose that, before looking at the data, we expected that the same amount of additional money matters more for people who generally have less money. This translates into a higher a-priori values of $$\zeta_1$$ (difference between ‘below_20’ and ‘20_to_40’) and hence into higher values of $$\alpha_1$$. We choose $$\alpha_1 = 2$$ and $$\alpha_2 = \alpha_3 = 1$$, the latter being the default value of $$\alpha$$. To fit the model we write:

prior4 <- prior(dirichlet(c(2, 1, 1)), class = "simo", coef = "moincome1")
fit4 <- brm(ls ~ mo(income), data = dat,
prior = prior4, sample_prior = TRUE)

The 1 at the end of "moincome1" may appear strange when first working with monotonic effects. However, it is necessary as one monotonic term may be associated with multiple simplex parameters, if interactions of multiple monotonic variables are included in the model.

summary(fit4)
 Family: gaussian
Links: mu = identity; sigma = identity
Formula: ls ~ mo(income)
Data: dat (Number of observations: 100)
Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
total post-warmup samples = 4000

Population-Level Effects:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
Intercept    31.09      1.55    28.03    34.07 1.00     2860     2831
moincome     14.76      0.70    13.40    16.10 1.00     2628     2787

Simplex Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
moincome1     0.65      0.04     0.57     0.73 1.00     3370     2523
moincome1     0.24      0.05     0.15     0.32 1.00     4009     2593
moincome1     0.12      0.04     0.04     0.19 1.00     3428     2200

Family Specific Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sigma     7.07      0.51     6.17     8.13 1.00     3674     2724

Samples were drawn 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).

We have used sample_prior = TRUE to also obtain samples from the prior distribution of simo_moincome1 so that we can visualized it.

plot(fit4, pars = "prior_simo", N = 3) As is visible in the plots, simo_moincome1 was a-priori on average twice as high as simo_moincome1 and simo_moincome1 as a result of setting $$\alpha_1$$ to 2.

dat$age <- rnorm(100, mean = 40, sd = 10) We are not only interested in the main effect of age but also in the interaction of income and age. Interactions with monotonic variables can be specified in the usual way using the * operator: fit5 <- brm(ls ~ mo(income)*age, data = dat) summary(fit5)  Family: gaussian Links: mu = identity; sigma = identity Formula: ls ~ mo(income) * age Data: dat (Number of observations: 100) Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1; total post-warmup samples = 4000 Population-Level Effects: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS Intercept 37.26 5.70 26.68 48.99 1.00 1507 1936 age -0.18 0.16 -0.50 0.11 1.00 1403 1767 moincome 10.59 2.36 5.88 15.11 1.00 1291 1778 moincome:age 0.11 0.06 -0.00 0.24 1.00 1229 1825 Simplex Parameters: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS moincome1 0.70 0.12 0.44 0.91 1.00 1522 1478 moincome1 0.20 0.11 0.02 0.42 1.00 1539 1851 moincome1 0.10 0.06 0.00 0.24 1.00 2201 1448 moincome:age1 0.45 0.23 0.03 0.87 1.00 1554 1867 moincome:age1 0.34 0.21 0.02 0.80 1.00 1859 2497 moincome:age1 0.21 0.16 0.01 0.62 1.00 2314 1953 Family Specific Parameters: Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS sigma 6.99 0.51 6.09 8.06 1.00 2871 2192 Samples were drawn using sampling(NUTS). For each parameter, Bulk_ESS and Tail_ESS are effective sample size measures, and Rhat is the potential scale reduction factor on split chains (at convergence, Rhat = 1). conditional_effects(fit5, "income:age") ## Modelling Monotonic Group-Level Effects Suppose that the 100 people in our sample data were drawn from 10 different cities; 10 people per city. Thus, we add an identifier for city to the data and add some city-related variation to ls. dat$city <- rep(1:10, each = 10)
var_city <- rnorm(10, sd = 10)
dat$ls <- dat$ls + var_city[dat\$city]

With the following code, we fit a multilevel model assuming the intercept and the effect of income to vary by city:

fit6 <- brm(ls ~ mo(income)*age + (mo(income) | city), data = dat)
summary(fit6)
 Family: gaussian
Links: mu = identity; sigma = identity
Formula: ls ~ mo(income) * age + (mo(income) | city)
Data: dat (Number of observations: 100)
Samples: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
total post-warmup samples = 4000

Group-Level Effects:
~city (Number of levels: 10)
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sd(Intercept)              18.91      5.14    11.36    31.07 1.00     1408     1981
sd(moincome)                1.94      1.10     0.15     4.39 1.01     1090     1371
cor(Intercept,moincome)    -0.50      0.38    -0.98     0.40 1.00     2787     2382

Population-Level Effects:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
Intercept       48.02      8.86    31.95    66.15 1.00     1313     1741
age             -0.32      0.18    -0.69    -0.00 1.00     1632     1755
moincome         9.44      2.61     4.10    14.33 1.00     1649     2063
moincome:age     0.15      0.07     0.02     0.29 1.00     1601     1567

Simplex Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
moincome1         0.71      0.14     0.39     0.93 1.00     2019     2249
moincome1         0.19      0.12     0.01     0.48 1.00     1794     2389
moincome1         0.09      0.06     0.00     0.24 1.00     3429     2675
moincome:age1     0.53      0.23     0.06     0.90 1.00     2297     2788
moincome:age1     0.32      0.20     0.02     0.77 1.00     2407     2566
moincome:age1     0.15      0.12     0.01     0.48 1.00     3199     2690

Family Specific Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sigma     6.25      0.51     5.37     7.39 1.00     2826     2899

Samples were drawn 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).

reveals that the effect of income varies only little across cities. For the present data, this is not overly surprising given that, in the data simulations, we assumed income to have the same effect across cities.