📣 **FOREWORD **

I used public data available in different formats from different sources within the City-Council Statistics dpt., the BCNOpenData Web and the Web Project Inside Airbnb to create this Datasets with the intention of representing yearly series of all 73 Barcelona city’s neighbourhoods including variables historically related to displacement and neighbourhood change phenomena.

BCN Dataset Features

• year: Year the data was taken, the range goes from 2015 to 2019
• neighbourhood: Each of the 73 neighbourhoods of Barcelona
• housing(m2): Area devoted to housing as per registered activity
• parking(m2): Area devoted to parking as per registered activity
• comerce(m2): Area devoted to commerce as per registered activity
• industry(m2): Area devoted to industry as per registered activity
• offices(m2): Area devoted to offices as per registered activity
• education(m2): Area devoted to education as per registered activity
• healthcare(m2): Area devoted to healthcare as per registered activity
• hostelry(m2): Area devoted to hostelry as per registered activity
• sports(m2): Area devoted to sports as per registered activity
• religious(m2): Area devoted to religious as per registered activity
• shows(m2): Area devoted to shows as per registered activity
• other uses(m2): Area devoted to other uses as per registered activity
• avg €/€/month: Average rent price per month
• avg €/m2 Average rent prices per squared meter
• avg housing(m2): Average Household size in squared meters
• new contracts 1000 hab: Number of New rent contracts per thousand inhabitants
• expired contracts 1000 hab: Number of Terminated rent contracts per thousand inhabitants
• win/lost rents 1000 hab: Difference of new minus terminated rent contracts per thousand inhabitants
• binary rent growth_1000_hab: Same as preceeding but in binary; 0 is negative rent growth whereas 1 represents positive rent growth
• population: Population censed in each neighbourhood
• % spaniards: Percentage of locals
• % strangers: Percentage of strangers
• % w/ higher education: Percentage of people with higher studies
• unemployed: Number of unemployed people
• gini index(%): A measure of statistical dispersion intended to represent the income inequality or the wealth inequality within a nation or a social group. (0 means perfect equality and 100 perfect inequality)
• disp income(€/year): Households disposable income average per capita. Meaning the average amount of money left over that every person per household has, after paying taxes (a.k.a Net Income).
• new household purchases: Number of registered purchases of new (less than five years old) households
• protected household purchases: Number of registered purchases of public housing households
• used household purchases: Number of registered purchases of used households
• new household purchases(x1000€): Average price (in thousands of €) of registered purchases of new (less than five years old) households
• Used household purchases(x1000€): Average price (in thousands of €) of registered purchases of used households
• Total household purchases(x1000€): Average price (in thousands of €) of total registered household purchases

The bcn_datasets can be conceptually divided in three categories:

1. Housing Factors: Here we include terms relative to the rent market as well as the real state market. From average prices to amount of registered purchases and a rent growth rate, calculated subtracting expired contracts from newly signed contracts. Also the area of each neighbourhood classified per registered activity, that provides a clearer sense of the "Urban-Scheme" for each of the studied neighbourhoods.

2. Resident Characteristics: Where we include percentages of population by certain conditions like higher education, foreigners and unemployed.

3. Economic Info: Here we include terms relative to wealth (i.e. disposable income) and it’s inequality distribution (i.e. Gini index).

The Airbnb dataset originally contained all features available from the listings in barcelona during the period 2009 - 2022 . All features where removed except for the basic ones related with the subject of this study. Like Price, License, Host id. , Flat id., First and Last review date, Neighbourhood and a few more.
The intention was to carry an EDA separately and at some point in the study merge both datasets in just one. Don't know when , though 🙂

🛑 BEFORE YOU START ✋

Information Publicly available was extracted from pdfs and csv, mostly, a thorough cleaning was carried due to the poor coherence of Data Entry over the years (mostly typos) but still data from rent prices for years 2012-2013 is missing. KNN imputation might do the trick in most cases.

Also I provide 2 different datasets:

• bcn_dataset_2009-2019
• bcn_dataset_2015-2019

The second one has more features since a National Public Study (Atlas de Distribucion de Renta de los Hogares) was available for those years, hence I found it relevant to include the features that weren't available from earlier years (e.g Gini Index).

Anyways the dataset is going to be improved within due time. 🗿

Enjoy! 🤸‍♂️