Preparation

Here is the raw data imported from the computer.

raw_df <- readxl::read_xlsx ("/Users/cucar/Downloads/EVDS_istanbul_property_data.xlsx")

Converting data to get reasonable results from plots

Monthly property sales data are selected with new headliners to get meaningful plots.

str(raw_df)
## tibble [187 × 9] (S3: tbl_df/tbl/data.frame)
##  $ Tarih                : chr [1:187] "2010-01" "2010-02" "2010-03" "2010-04" ...
##  $ TP AKONUTSAT1 T40    : chr [1:187] NA NA NA NA ...
##  $ TP AKONUTSAT2 T40    : chr [1:187] NA NA NA NA ...
##  $ TP AKONUTSAT3 T40    : num [1:187] NA NA NA NA NA NA NA NA NA NA ...
##  $ TP AKONUTSAT4 T40    : num [1:187] NA NA NA NA NA NA NA NA NA NA ...
##  $ TP DISKONSAT ISTANBUL: num [1:187] NA NA NA NA NA NA NA NA NA NA ...
##  $ TP HEDONIKYKFE IST   : num [1:187] 35.9 36.6 37.4 38 38 37.6 37.3 38.1 38.8 39.1 ...
##  $ TP HKFE02            : num [1:187] 36 36.2 36.5 36.9 37.1 37 37.2 37.3 37.7 38.1 ...
##  $ TP TCBF02 ISTANBUL   : num [1:187] 1415 1420 1428 1443 1449 ...
sales <- slice(raw_df, 37:129) %>% select(date='Tarih', total_property_sales='TP AKONUTSAT1 T40', mortgage_sales= 'TP AKONUTSAT2 T40',  sale_of_new_properties= 'TP AKONUTSAT3 T40', sale_of_old_properties= 'TP AKONUTSAT4 T40')
sales$date <- readr::parse_date(sales$date, format ="%Y-%m") 
sales$total_property_sales <- as.integer(sales$total_property_sales)
sales$mortgage_sales <- as.integer(sales$mortgage_sales)
sales$sale_of_new_properties <- as.integer(sales$sale_of_new_properties)
sales$sale_of_old_properties <- as.integer(sales$sale_of_old_properties)
ggplot(sales, aes(x = date , y = mortgage_sales, color=total_property_sales)) +geom_point() + theme_minimal()

From the scatter plot above, we can say that, mortgage sales were between 5,000 and 10,000 until the pandemic. When the economy went bad during the beginning of pandemic, mortgage sales fell around 1,000. Then, despite increase in total property sales which is around 40,000, mortgage sales came up to around 15,000 - 20,000.

Index data are selected with new headliners to get reasonable results.

index_data <- raw_df %>% select(date='Tarih', new_property_price_index= 'TP HEDONIKYKFE IST', property_price_index= 'TP HKFE02', price_per_m2_try= 'TP TCBF02 ISTANBUL')
index_data <- slice(index_data, 0:128)
index_data$date <- readr::parse_date(index_data$date, format ="%Y-%m")

Next;

Code preparation as below;

index_data <- raw_df %>% select(date='Tarih', new_property_price_index= 'TP HEDONIKYKFE IST', property_price_index= 'TP HKFE02', price_per_m2_try= 'TP TCBF02 ISTANBUL')
index_data <- slice(index_data, 0:128)
index_data$date <- readr::parse_date(index_data$date, format ="%Y-%m") 
ggplot(index_data, aes(x = date, y = property_price_index, color=price_per_m2_try)) +geom_point() + theme_minimal()

From this scatter plot, we conclude that, after 2019 in Istanbul (after the pandemic), average price increase for a property has increased more regarding Turkey’s average price increase for a property.