% Author: Simon-Pierre Boucher — contact@spboucher.ai
%
\begin{table}[!htbp]
\centering
\caption{Comparison of text representation methods in the hedonic model. $\Delta R^2$ is the improvement over Model~A (structural only, $R^2 = 0.452$).}
\label{tab:text_comparison}
\small
\begin{adjustbox}{max width=\textwidth}
\begin{tabular}{lR{1.5cm}C{2.2cm}C{2.6cm}}
\toprule
\textbf{Text Representation} & $\boldsymbol{\Delta R^2}$ & \textbf{Interpretability} & \textbf{Economic Usefulness} \\
\midrule
Description length only & +0.012 & High & Low \\
PCA on embeddings (20 PC) & +0.070 & None & Low \\
PCA on embeddings (20 PC) + length & +0.078 & None & Low \\
Reference cosine (20 dim) & +0.044 & High & High \\
\textbf{Reference cosine (20 dim) + length} & \textbf{+0.060} & \textbf{High} & \textbf{High} \\
\bottomrule
\multicolumn{4}{p{0.95\textwidth}}{\footnotesize{\textit{Notes:} All specifications include the six structural variables and are estimated on the same $n = 17{,}087$ sample. The 20 principal components capture 49.5\% of the variance of the raw 384-dimensional embeddings. PCA achieves a larger fit improvement than the reference projections but produces features with no economic interpretation; the reference-based approach trades roughly two percentage points of $R^2$ for coefficient-level interpretability (see Section~\ref{sec:robustness}).}} \\
\end{tabular}
\end{adjustbox}
\end{table}