Figure 1. Comparison of house price growth in Shelby County and the USA (2003–2023) Note(s): This figure illustrates the historical trend of house price growth in Shelby County, TN, versus the house price growth in the USA over the past 21 years. Blue shaded region, showing the last bubble.... More about this image found in Comparison of house price growth in Shelby County and the USA (2003–2023)...

Figure 2. Trends in house price growth and key economic indicators (2006–2022) Note(s): This figure illustrates the dynamic relationship between house price growth (blue bars) and demand fundamentals (orange line) from the first quarter of 2006 to the fourth quarter of 2022. Panel (a) illustrat... More about this image found in Trends in house price growth and key economic indicators (2006–2022) Note...

Figure 3. Trends in sales turnover and house price growth (2006–2022) Note(s): This figure illustrates the dynamic relationship between house price growth (blue bars) and sales turnover (orange line) from the first quarter of 2006 to the fourth quarter of 2022. The sales turnover rate is calculated as the number of monthly sales contracted divided by the number of houses listed for sale in the same period. The left vertical axis represents changes in house prices, while the right vertical axis indicates percentage changes in sales turnover A combo chart compares house price growth bars with sales turnover from 2006 Quarter 1 to 2022 Quarter 3. The combo chart presents House Price Growth as bars and Sales Turnover as a line from 2006 Quarter 1 to 2022 Quarter 3. The left vertical axis ranges from negative 1.5 to 2. The right vertical axis ranges from 0 to 1. House Price Growth is mostly positive in 2006, negative from 2007 to 2011, mostly positive from 2012 onward, highest around 2020 to 2021, and slightly negative near 2022 Quarter 3. Sales Turnover starts near 0.6 in 2006, falls below 0.5 around 2007 to 2008, rises gradually after 2011, reaches about 0.9 around 2020 to 2021, and falls to about 0.7 in 2022. More about this image found in Trends in sales turnover and house price growth (2006–2022) Note(s): Thi...

Figure 1. Conceptual framework Note(s): H1: proximal theory-of-planned-behaviour predictors of intention; H2: mid-level mechanisms (asset logic, risk perception, community amenities) shaping attitude, perceived behavioural control and intention; H3: distal cognitive and informational drivers (optimism bias, flood information transparency). Multigroup analysis compares recent residents with long-term residents and renters with owner-occupiers A conceptual model links distal drivers, mid-level mechanisms, proximal T P B factors, and purchase intention for flood-related housing decisions. The model is organised into 3 sections. The H 3 Distal drivers section contains Optimism bias, O B, and Flood information transparency, F I T. Optimism bias points to Housing bullishness, H B, through H 3 a; Perceived flood risk, P F R, through H 3 b; and Flood information transparency through H 3 c. Flood information transparency points to Housing bullishness through H 3 d and Perceived flood risk through H 3 e. The H 2 Mid-level mechanisms section contains Housing bullishness, Perceived flood risk, and Community convenience, C C. Housing bullishness points to Attitude, A T T, through H 2 a. Perceived flood risk points to Attitude through H 2 b, Perceived behavioural control, P B C, through H 2 c, and Subjective norm, S N, through H 1 d. Community convenience points to Attitude through H 2 d, Purchase intention, P I, through H 2 e, and Perceived behavioural control through H 2 f. Flood information transparency also points to Attitude through H 2 g and Perceived behavioural control through H 2 h. The H 1 Proximal T P B section contains Attitude, Subjective norm, Perceived behavioural control, and Purchase intention. Attitude points to Purchase intention through H 1 a. Subjective norm points to Purchase intention through H 1 b. Perceived behavioural control points to Purchase intention through H 1 c. A multigroup analysis compares Recent Residents with Long-Term Residents and Rent House with Buy House. More about this image found in Conceptual framework Note(s): H1: proximal theory-of-pla...

Figure 1. Indexed Price Trends of MSP, OHP and Adjusted OLP across Policy Regimes (2018 = 100) Note(s): MSP denotes market sale prices, OHP officially assessed housing prices and adjusted OLP officially assessed land prices multiplied by a standardized land share of 40 m². The vertical lines indicate the 2019 assessed value reform and the 2023 policy retreat period A line graph compares market price, administrative housing price, and adjusted online listing price indices over time. The line graph plots index with 2018 equal to 100 on the vertical axis against year from 2006 to 2025 on the horizontal axis. The legend identifies M S P, market price, O H P, administrative housing price, and adjusted O L P, O L P times 40. M S P remains near 62 to 64 until 2014, rises to 100 in 2018, increases to about 146 in 2021, declines to about 129 in 2023, and recovers to about 163 in 2025. O H P rises from about 61 in 2006 to about 72 in 2007, remains near 71 to 73 until 2014, reaches 100 in 2018, increases to about 160 in 2022, falls to about 129 in 2023, and rises to about 145 in 2025. Adjusted O L P increases from about 56 in 2006 to about 75 in 2011 to 2014, reaches 100 in 2018, rises to about 163 in 2022, remains above 150 after 2023, and ends near 164 in 2025. Two vertical reference lines mark the 2019 reform and the 2023 retreat. More about this image found in Indexed Price Trends of MSP, OHP and Adjusted OLP across Policy Regimes...

Figure 2. Growth Gap between OHP and OLP (Amplification Effect) Note(s): The figure plots the difference between the growth rates of officially assessed housing prices ( OHP ) and adjusted land prices (OLP × 40). Positive values indicate that OHP grows faster than OLP . Vertical lines de... More about this image found in Growth Gap between OHP and OLP (Amplification Effect) Note(s): The f...

Figure 3. Growth Gap between OHP and MSP (Growth Reversal) Note(s): The figure plots the difference between the growth rates of officially assessed housing prices ( OHP ) and market sale prices ( MSP ). Positive values indicate that OHP grows faster than MSP, suggesting a temporary growth... More about this image found in Growth Gap between OHP and MSP (Growth Reversal) Note(s): The figure...

Figure 4. Asymmetric Adjustment of OHP Across Market Conditions Note(s): The figure compares the average growth rate of officially assessed housing prices ( OHP ) under different market conditions. Market conditions are defined based on the sign of MSP growth. The results show that OHP in... More about this image found in Asymmetric Adjustment of OHP Across Market Conditions Note(s): The fig...

Figure 1. Pearson correlation matrix of property features and price (n = 59,035). The strongest linear association is between price and bathrooms (r = 0.35); bedrooms and bathrooms exhibit high collinearity (r = 0.94) A heat map shows Pearson r values for 7 property variables, with the strongest, 0.94, between Bedrooms and Bathrooms. The heat map compares Pearson r values for Price, Area, Bedrooms, Bathrooms, Type, Purpose, and Location. The diagonal values are 1.00 for each variable. Price has 0.21 with Area, 0.15 with Bedrooms, 0.35 with Bathrooms, negative 0.12 with Type, negative 0.15 with Purpose, and negative 0.08 with Location. Area has 0.04 with Bedrooms, 0.04 with Bathrooms, 0.17 with Type, negative 0.06 with Purpose, and 0.01 with Location. Bedrooms has 0.94 with Bathrooms, negative 0.37 with Type, 0.06 with Purpose, and negative 0.01 with Location. Bathrooms has negative 0.37 with Type, 0.00 with Purpose, and 0.04 with Location. Type has negative 0.06 with Purpose and 0.00 with Location. Purpose has negative 0.06 with Location. The scale labelled Pearson r runs from negative 1.00 to 1.00. More about this image found in Pearson correlation matrix of property features and price (n...

Figure 2. SHAP-based feature importance: mean |SHAP value| ( PKR ) for each predictor computed on a 5,000-row sample of the test set, tuned XGBoost A bar chart ranks S H A P feature importance, with Purpose Sale or Rent highest and Property type lowest. The bar chart is titled Feature importan... More about this image found in SHAP-based feature importance: mean |SHAP value| ( PKR ) for each predictor...

Figure 3. SHAP beeswarm summary plot. Each point is a single test observation; horizontal position is the SHAP value ( PKR contribution to predicted price); color encodes the feature value (red = high, blue = low) A S H A P summary plot ranks 6 features, with Purpose, Sale or Rent, Area in... More about this image found in SHAP beeswarm summary plot. Each point is a single test observation; horiz...

Figure A1. Comprehensive property market analysis – Islamabad database (n = 59,035) Note(s): The figure originally referenced as Section 4.9 in the previous submission is reproduced here as a single 12-panel visualization of the Islamabad property market. The panels encompass (a) price distribution; (b) property type distribution; (c) area vs price scatter; (d) top locations by count; (e) average price by bedrooms; (f) XGBoost feature importance; (g) actual vs predicted prices; (h) model R² comparison; (i) purpose distribution; (j) properties by price range; (k) average price by bathrooms; and (l) price per Marla distribution. The figure file ( Figure A1 _Comprehensive_Panel.png) is supplied separately as part of the submission package A twelve-panel dashboard presents property price, type, location, model, purpose, bedroom, bathroom, and price-per-Marla analyses. The dashboard has 12 panels labelled a to l. Panel A is a Price Distribution histogram with Frequency on the vertical axis and Price in P K R million on the horizontal axis from 0 to 1,000. A median marker appears near the lower price range. Panel b is a Property Type Distribution bar chart with Number of Listings on the vertical axis. House has 28,842 listings. Plot has 14,195. Flat has 7,315. Upper Portion has 3,082. Lower Portion has 2,232. Other has 3,369. Panel c is an Area versus Price scatter plot. Area in Marla runs from 0 to 120. Price in P K R million runs from 0 to about 150. Points rise as area increases. Panel d lists the Top 5 Locations by Count. Bahria Town Phase 8 is about 3,200 listings. D H A Defence is about 2,500. D H A Phase 6 is about 2,100. D H A Phase 5 is about 1,400. Bahria Town R W P is about 1,300. Panel e plots Average Price by Bedrooms. The average price rises from about 9 P K R million for 1 bedroom to about 78 P K R million for 6 bedrooms. Panel f is X G Boost Feature Importance. Bathrooms is about 32 per cent. Purpose is about 30 per cent. Area is about 13 per cent. Type is about 11 per cent. Location is about 9 per cent. Bedrooms is about 5 per cent. Panel g is Actual versus Predicted Prices for the X G Boost test set. Actual Price and Predicted Price are in P K R million, and both axes run from 0 to about 200. A perfect fit line crosses the plot. Panel h compares model R squared scores. X G Boost is 0.695. Random Forest is 0.692. Decision Tree is 0.579. Gradient Boosting is 0.552. Linear regression is 0.251. S V R is negative 0.069. Panel i is Purpose Distribution. For Sale is 78.3 per cent. For Rent is 21.3 per cent. Other is 0.4 per cent. Panel j is Distribution by Price Range. Less than 10 million is 30 per cent. 10 to 30 million is 35 per cent. 30 to 50 million is 15 per cent. More than 50 million is 20 per cent. Panel k plots Average Price by Bathrooms, called Quality Gradient Effect. The average price rises from about 9 P K R million for 1 bathroom to about 135 P K R million for 7 bathrooms. Panel l is a Distribution of Price per Marla histogram. Price per Marla in P K R million runs from 0 to about 15, and Frequency reaches about 500. More about this image found in Comprehensive property market analysis – Islamabad database (n...

Figure 1. Factors influencing flood susceptibility Source: Authors’ own work A flow diagram presents six flood susceptibility factors leading to flood susceptibility. The flow diagram presents six inputs leading to flood susceptibility. The inputs are elevation, slope, land use and land cov... More about this image found in Factors influencing flood susceptibility Source: Authors’ own work A ...

Figure 2. Hierarchical structure of MCDM techniques Source: Authors’ own work A flowchart classifies M C D M into M C D A and M O D M, with M C D A methods grouped by type. The flowchart presents M C D M classification. M C D M branches into M C D A and M O D M. M C D A branches into rank... More about this image found in Hierarchical structure of MCDM techniques Source: Authors’ own work ...

Figure 3. Research framework Source: Authors’ own work A flowchart presents phases for flood risk development, G I S integration, and flood-related map outputs. The flowchart presents four phases and a final output. Phase 1 starts with a case study, Taman Sri Muda, Shah Alam. Phase 2 covers... More about this image found in Research framework Source: Authors’ own work A flowchart presents pha...