brand_buyer_nu#

pymc_marketing.customer_choice.taste_profiles.brand_buyer_nu(model, n_samples=200, dim=0)[source]#

Posterior-mean buyer taste per brand and market, for one taste dimension.

For each market \(m\) and brand \(j\),

\[\bar\nu_{m,j,d} = E[\nu_d \mid \mathrm{buys\ brand\ } j \mathrm{\ in\ } m]\]

averaged across posterior draws.

Parameters:

modelBayesianBLP: A fitted model.
n_samplesint: Number of posterior draws.
dimint: Which random-coefficient dimension to slice (0 is typically price).

Returns:

np.ndarray: Shape (M, J). Posterior-mean buyer taste on the chosen dimension for each (market, brand).

Raises:

IndexError: If dim is outside [0, n_random).