Bayesian workshop - STEP 2023
University of Alberta
Using a Normal distribution to model reaction times:
Most models in linguistics model changes in the mean only
The distribution moves up or down based on another predictor
rt
coming from Normal distribution with mean (\(\mu\)) and standard deviation (\(\sigma\)) Frequency
(Conditional mean)rt
equal Frequency
here?
\(rt \sim Normal(\mu,\sigma)\)
\(\mu = Frequency\)
Frequency
and rt
rt
does not change with Frequency
rt
\(\uparrow\) when Frequency
\(\uparrow\)rt
\(\downarrow\) when Frequency
\(\uparrow\)rt
changes\(rt \sim Normal(\mu,\sigma)\)
\(\mu = \beta_1Frequency\)
Frequency
= 0?rt
does not change with Frequency
rt
\(\uparrow\) when Frequency
\(\uparrow\)rt
\(\downarrow\) when Frequency
\(\uparrow\)rt
changes\(rt \sim Normal(\mu,\sigma)\)
\(\mu = \beta_1Frequency\)
We add another parameter: Intercept
This is the value of rt
when Frequency
= 0
\(rt \sim Normal(\mu,\sigma)\)
\(\mu = Intercept + \beta_1 Frequency\)
This is how Bayesian models are often written. We specify a model, then specify priors for all parameters
Model:
\(rt \sim Normal(\mu,\sigma)\)
\(\mu = Intercept + \beta_1 Frequency\)
Priors:
\(Intercept \sim Normal(?,?)\)
\(\beta_1 \sim Normal(?,?)\)
\(\sigma \sim Normal_+(?,?)\)
Intercept:
- Prior belief on what rt
could be when Frequency
= 0
\(\beta_1\):
- Prior belief on how rt
changes when Frequency
increases by 1
\(\sigma\):
- Prior belief on how variable reaction times are
Model:
\(rt \sim Normal(\mu,\sigma)\)
\(\mu = Intercept + \beta_1 Frequency\)
Priors:
\(Intercept \sim Normal(?,?)\)
\(\beta_1 \sim Normal(?,?)\)
\(\sigma \sim Normal_+(?,?)\)
rt
by Frequency
Open the script entitled S2_E1_linear_model.R