spb/wp11_uqo Public

1% Author: Simon-Pierre Boucher — contact@spboucher.ai2% ============================================================================3\section{Data}4\label{sec:data}56\subsection{Sources}78We use a registry of 745{,}119 Quebec residential-market transactions9recorded between January 2021 and July 2026 --- sale price, date, address10and coordinates --- each matched at the parcel level to the municipal11assessment roll in force at the sale date (median match distance120.6~metres, validated against the roll's recorded value). The roll13contributes the structural descriptors that are notoriously missing from14transaction registries: total floor area, lot area, year of construction,15number of dwelling units, number of storeys, physical configuration16(detached, semi-detached, row, apartment-linked) and the standardized use17code. This combination --- market prices with assessor-grade structure18for half a million sales across an entire province --- is what makes a19controlled specification race feasible: every model sees exactly the same20attributes.2122\subsection{Sample}2324We keep residential use codes (dwellings, cottages, mobile homes), require25a high-confidence roll match (distance $\le 50$~m, score $\ge 150$), a26price of at least \$50{,}000, a plausible floor area (20--2{,}000~m$^2$)27and a known year of construction, and trim prices and floor areas at the281st and 99th percentiles within each sale year. The estimation sample29contains \textbf{514{,}212 sales}: 348{,}253 single-family homes, 75{,}73430condominiums, 74{,}909 plexes (2--5 units), 10{,}349 cottages and 4{,}88131mobile homes, spread over 1{,}115 municipalities, 5{,}255 grid cells of32$\sim$5.5~km and 30{,}548 cells of $\sim$1.1~km.33Table~\ref{tab:sumstats} reports the descriptives. Prices average34\$462{,}000 (median \$399{,}000); the window covers the tail of the35post-pandemic boom, the 2022--2023 rate correction and the subsequent36recovery --- a demanding environment for any static price model, which we37exploit in the forward-in-time evaluation.3839\begin{table}[t]40\centering41\begin{threeparttable}42\caption{Summary statistics, estimation sample}43\label{tab:sumstats}44\small45\input{../results/tables/summary_stats}46\begin{tablenotes}[flushleft]\footnotesize47\item \textit{Notes:} 514{,}212 residential sales, January 2021 -- July482026. Structural descriptors from the assessment roll in force at the49sale date. Lot area is zero for condominiums and other units without an50exclusive lot.51\end{tablenotes}52\end{threeparttable}53\end{table}54