Longitudinal modelling of housing prices with machine learning and temporal regression
Bibliographic record
Abstract
Purpose The purpose of this paper is to model housing price temporal variations and to predict price trends within the context of land use–transportation interactions using machine learning methods based on longitudinal observation of housing transaction prices. Design/methodology/approach This paper examines three machine learning algorithms (linear regression machine learning (ML), random forest and decision trees) applied to housing price trends from 2001 to 2016 in the Greater Toronto and Hamilton Area, with particular interests in the role of accessibility in modelling housing price. It compares the performance of the ML algorithms with traditional temporal lagged regression models. Findings The empirical results show that the ML algorithms achieve good accuracy (R2 of 0.873 after cross-validation), and the temporal regression produces competitive results (R2 of 0.876). Temporal lag effects are found to play a key role in housing price modelling, along with physical conditions and socio-economic factors. Differences in accessibility effects on housing prices differ by mode and activity type. Originality/value Housing prices have been extensively modelled through hedonic-based spatio-temporal regression and ML approaches. However, the mutually dependent relationship between transportation and land use makes price determination a complex process, and the comparison of different longitudinal analysis methods is rarely considered. The finding presents the longitudinal dynamics of housing market variation to housing planners.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".