Improve deciles for ratio curves

provisional-ratio-curves/provisional-ratio-curves.R

@@ -87,14 +87,17 @@ walk(unique(all_ratios$township_name), \(x) {
     filter(township_name == x) %>%
     select(-township_name) %>%
     mutate(
-      price_decile = ntile(sale_price, 10),
+      price_decile = ntile(
+        if_else(sale_excluded %in% TRUE, NA_real_, sale_price),
+        10
+      ),
       model_sale_ratio = model_value / sale_price,
       desk_review_sale_ratio = desk_review_value / sale_price
     )
 
   # Summarize ratios and price ranges by decile for .xlsx output
   output$`Ratio Deciles` <- output$`All Parcels` %>%
-    filter(!is.na(sale_price)) %>%
+    filter(!is.na(sale_price), !sale_excluded) %>%
     summarize(
       model_ratio = median(model_sale_ratio),
       model_cod = assessr::cod(model_sale_ratio),
@@ -114,7 +117,7 @@ walk(unique(all_ratios$township_name), \(x) {
 
   # Calculate vertical equity metrics for .xlsx output
   output$`Town-Level Stats` <- output$`All Parcels` %>%
-    filter(!is.na(sale_price)) %>%
+    filter(!is.na(sale_price), !sale_excluded) %>%
     summarize(
       model_ratio = median(model_sale_ratio),
       model_cod = assessr::cod(model_sale_ratio),
@@ -153,7 +156,7 @@ walk(unique(all_ratios$township_name), \(x) {
 
   # Build x-axis labels with abbreviated price ranges per decile
   axis_labels <- output$`All Parcels` %>%
-    filter(!is.na(`Sale Price`)) %>%
+    filter(!is.na(`Sale Price`), !`Sale Excluded`) %>%
     summarize(
       min_price = min(`Sale Price`),
       max_price = max(`Sale Price`),
@@ -229,6 +232,7 @@ walk(unique(all_ratios$township_name), \(x) {
     select(town_nbhd, geometry)
 
   map_data <- output$`All Parcels` %>%
+    filter(!isTRUE(`Sale Excluded`)) %>%
     select(
       `Neighborhood Number`,
       `Model Sale Ratio`,